SESAM USER MANUAL

WADAM
Wave Analysis by Diffraction and Morison theory

Valid from program version 9.3

SAFER, SMARTER, GREENER

Sesam User Manual

Wadam

Date: March/2017

Valid from program version 9.3

Prepared by DNV GL - Software

E-mail support: [email protected]

E-mail sales: [email protected]

c DNV GL AS. All rights reserved

This publication or parts thereof may not be reproduced or transmitted in any form or by any means, including copying
or recording, without the prior written consent of DNV GL AS.
Table of contents

1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1 Wadam – Wave Analysis by Diffraction and Morison Theory . . . . . . . . . . . . . . . . . . . . . 2
1.2 Wadam in the Sesam system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 How to Read this Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Terminology and Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Status List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.6 Wadam Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 FEATURES OF WADAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1 Definition of Model Types in Wadam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.1 The Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.2 The Panel Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.3 The Morison Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.1.4 The Composite Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.5 Single Super element Composite model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.6 Multi-Body Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.7 Mass Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.1.8 Structural Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.9 Free Surface Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.1.10 Load Case Numbering and Load Case Combinations . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2 Global Response Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2.2 Computation of Wave Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2.3 The Global Response Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.3 The Calculation of Detailed Loads on a Structural Model . . . . . . . . . . . . . . . . . . . . . . 24
2.3.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.3.2 The Structural Load Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.3.3 Deterministic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.4 Detailed Loads Transfer to a Model with Shell or Solid Elements . . . . . . . . . . . . . . . . . 25
2.3.5 Detailed Loads Transfer to a Model with Beam Elements . . . . . . . . . . . . . . . . . . . . . . 25
2.4 Environmental Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.4.1 Surface Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.4.2 Current Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.4.3 Water Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.5 Results Types Reported from Wadam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.5.1 Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.5.2 Result Reference Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.5.3 Dimensioning of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.5.4 Transfer Functions and Phase Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.5.5 Hydrostatic Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.5.6 Global Mass Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.5.7 Added Mass Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.5.8 Damping Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.5.9 Exciting Forces and Moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.5.10 Rigid Body Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.11 Second Order Mean Drift Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.12 Second Order Sum and Difference Frequency Results . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.13 Fluid Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.14 Wave Drift Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.15 Distributed Hydrostatic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.5.16 Distributed Hydrodynamic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.5.17 Load Sum Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.5.18 Sectional Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.5.19 Roll Damping Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.5.20 Global drag-coefficient for roll-damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.6 Calculation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.6.1 Calculation of Wave Loads from Potential Theory . . . . . . . . . . . . . . . . . . . . . . . . . . 37

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page i

 2.6.2 Calculation of Wave Loads from Second Order Potential Theory . . . . . . . . . . . . . . . . . . 38
2.6.3 Removal of Irregular Frequencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.6.4 Damping free surface lid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.6.5 The Equation of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.6.6 Morison’s Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.6.7 Calculation of Tank Pressures, Quasi-static Method . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.6.8 Calculation of Tank Pressures, Full Dynamic Method . . . . . . . . . . . . . . . . . . . . . . . . 42
2.6.9 Pressure Loads up to Free Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.6.10 Reduced pressure up to the free surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.6.11 Wave-current interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.7 The Save-Restart System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

3 USER’S GUIDE TO WADAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

3.1 Column-stabilized Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.2 Global Response for a Ship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.3 Handling of fluid compartments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.4 Free surface lids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

4 EXECUTION OF WADAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.1 Program Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.1.1 The Input Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.1.2 Output Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.1.3 The Save-Restart File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.2 Processor dependent optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.3 Program Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.4 Program Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.5 Warnings and Error Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

A THEORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A1
A1 Hydrostatic Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A1
A 1.1 Hydrostatic Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A1
A2 Morison Element Formulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A1
A 2.1 The Anchor Element Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A2
A 2.2 The TLP Mooring Element Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A2
A3 Calculation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A3
A 3.1 Linearisation of Roll Restoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A3
A 3.2 Calculation of Line Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A4
A 3.3 The Mapping of Loads from Panel Models to Finite Element Models . . . . . . . . . . . . . . . . A5
A 3.4 Calculation of Tank Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A6
A 3.5 Global drag-coefficient for roll . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7

B The Wadam Print File List of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B1

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 1

1 INTRODUCTION

1.1 Wadam – Wave Analysis by Diffraction and Morison Theory

Wadam is a general analysis program for calculation of wave-structure interaction for fixed and floating
structures of arbitrary shape, e.g. semi-submersible platforms, tension-leg platforms, gravity-base struc-
tures and ship hulls.

The analysis capabilities in Wadam comprise:

• Calculation of hydrostatic data and inertia properties

• Calculation of global responses including:
– First and second order wave exciting forces and moments

– Hydrodynamic added mass and damping

– First and second order rigid body motions

– Sectional forces and moments

– Steady drift forces and moments

– Wave drift damping coefficients

– Internal tank pressures

• Calculation of selected global responses of a multi-body system

• Automatic load transfer to a finite element model for subsequent structural analysis including:
– Inertia loads

– Line loads for structural beam element analysis

– Pressure loads for structural shell/solid element analysis

– Pressure loads up to the free surface

Wadam calculates loads using:

• Morison’s equation for slender structures

• First and second order 3D potential theory for large volume structures
• Morison’s equation and potential theory when the structure comprises of both slender and large
volume parts. The forces at the slender part may optionally be calculated using the diffracted wave
kinematics calculated from the presence of the large volume part of the structure.
The Wadam results may be presented directly as complex transfer functions or converted to time domain
results for a specified sequence of phase angles of the incident wave. For fixed structures Morison’s equa-
tion may also be used with a time domain output option to calculate drag forces due to time independent
current.

The same analysis model may be applied to both the calculation of global responses in Wadam and the
subsequent structural analysis. For shell and solid element models Wadam also provides automatic mapping
of pressure loads from a panel model to a differently meshed structural finite element model.

The 3D potential theory in Wadam is based directly on the Wamit program developed by Massachusetts
Institute of Technology Ref. [1] and Ref. [2].

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 2

1.2 Wadam in the Sesam system

Wadam is an integrated part of the Sesam suite of programs. It is tailored to calculate wave loads on models
created by the Sesam preprocessors Patran-Pre, Prefem, GeniE and Presel. The models are read by Wadam
from the Input Interface File (T-file). The Wadam analysis control data is generated by the Hydrodynamic
design tool HydroD.
The results from the Wadam global response analysis may be stored on a Hydrodynamic Results Interface
File (G-file) for statistical postprocessing in Postresp. The loads mapped to structural finite elements may be
stored on the Loads Interface File (L-file) for a subsequent structural analysis in Sestra.

Figure 1.1 shows Wadam in the Sesam system. A detailed description of the input and output files is given
in Section 4.

Figure 1.1: Sesam overview

1.3 How to Read this Manual

Section 2 FEATURES OF WADAM describes the problems Wadam can solve. Descriptions of models, environ-
ment and results produced by Wadam are included.

Section 3 USER’S GUIDE TO WADAM presents tutorial examples. Each example includes a discussion of the
modelling, execution and results interpretation phases. Both simple examples and engineering applications
are included.

Section 4 EXECUTION OF WADAM describes the input and output files of Wadam. The memory and disk
storage requirements are discussed together with some rules of thumb on execution time for different types
of analysis. Finally, the chapter lists the problem size limitations in Wadam.

Appendix A THEORY includes additional theory description for Wadam.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 3

1.4 Terminology and Notation

Composite model A hydro model consisting of a panel model and Morison model repre-
senting separate parts of the structure. See Section 2.1.4.
FE Abbreviation for ’finite-element’ like in FE model.

HydroD The hydrodynamic design tool. This is a modern graphical tool for
modelling the input data to Wadam. Wadam is started directly from
within this application.
Hydrodynamic Results Interface File A file containing results from a global response analysis. This file is
termed G-file for short. The post-processor Postresp performs sta-
tistical post-processing of these results. See Section 2.2 and Sec-
tion 4.1.2.

Hydro model Hydrodynamic model. A model used for calculating hydrodynamic

loads from potential theory and Morison’s equation. See Section 2.1.

Hydro property Hydrodynamic properties including added mass, drag coefficients, el-
ement diameters and anchor characteristics required for calculating
hydrodynamic loads.

Input Interface File A file containing geometrical information of the structure (FE model
plus hydro model). This file is termed T-file for short. See Sec-
tion 4.1.1.

Loads Interface File A file containing loads for a subsequent structural analysis. This file
is termed L-file for short. See Section 4.1.2.

Prewad Was Wadam’s interactive preprocessor. This has now been removed.
RSQ File Wadam’s save-restart file <user-defined name>.RSQ; see Section 2.7
and Section 4.1.3.
S-file A file containing information on the relation between load cases and
wave frequencies. See Section 2.3.
swl Abbreviation for ’still water level’.

1.5 Status List

There exists for Wadam (as for all other Sesam programs) a Status List providing additional information. This
may be:

• Reasons for update (new version)

• New features
• Errors found and corrected
• Etc.
The most recently updated status lists can be accessed over the internet. Go to www.dnvgl.com/software
and select the Support link. Then log in to the Customer portal for Sesam. A user name and password is
required to access this site. Please follow the instructions given to obtain access.

In the Status List Browser window narrow the number of entries listed:

• Entries relevant to a specific version only

• Entries of a specific type, e.g. Reasons-for-Update
• Entries containing a given text string

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 4

1.6 Wadam Extensions

Wadam has the following extensions which require separate license keys:

2ORD Calculation of second order sum- and difference- frequency transfer functions for bodies in
monochromatic and bi-chromatic incident waves.

NBOD Calculation of first-order hydrodynamic analysis of multiple bodies.

STRU Transfer of hydrodynamic loads to structural finite element models with beam, shell and solid ele-
ments.
WDD Computation of Wave Drift damping coefficients.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 5

2 FEATURES OF WADAM

This section describes the features of the Wadam program. The section is organised with Section 2.1 de-
scribing the modelling concepts adopted in Wadam. Thereafter, Section 2.2 and Section 2.3 describe the
two main analysis capabilities:
• Global response analysis for calculating rigid body type of results.
• Detailed load calculation for transfers of finite element type of loads to a structural model.
The environmental definition including both surface waves and current profiles is introduced in Section 2.4.
The description of the results produced by Wadam is included in Section 2.5. Basic calculation methods are
described in Section 2.6. Additional information on calculation methods is included in Appendix A.

2.1 Definition of Model Types in Wadam

The definition of models in Wadam includes three main model types: (1) the hydro model which is used to
calculate hydrodynamic forces, (2) the structural model where hydrodynamic and hydrostatic loads are rep-
resented as finite element loads and (3) the mass model. The mass model is relevant for floating structures
only and may be defined as finite elements with mass properties or as a global mass matrix. Alternatively,
the distributed mass points specified in a file can be given as input to the mass model from HydroD. The
different model types, see Figure 2.1, are described in this section. Wadam reads the various models from
Input Interface Files generated by e.g. the Sesam preprocessors GeniE, Patran-Pre and Presel.

Figure 2.1: Overview of model types in Wadam

The hydro model may, see Figure 2.1, consist of either:

• A panel model for calculation of hydrodynamic results from potential theory

• A Morison model for calculation of hydrodynamic loads from Morison’s equation
• A combination of a panel- and a Morison model – called a composite model. The composite model is
used when potential theory and Morison’s equation are applied to different parts of the hydro model.
The legal combinations of panel and Morison models are shown in Figure 2.2.

The element types interpreted by Wadam are listed in Table 2.1. Other element types present in the various
Wadam models will be neglected.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 6

 Figure 2.2: Hydro model combinations

Table 2.1: Overview of Sesam element types interpreted in the various Wadam models

Number Panel Morison Mass Structural

Type of element of nodes model model model model
Beam elements
Beam 2 v v v
Curved beam 3 v v
Shell elements
Triangular flat thin shell 3 v v v
Quadrilateral flat thin shell 4 v v v
Subparametric curved triangular thick shell 6 v v v
Subparametric curved quadrilateral thick shell 8 v v v
Multilayered curved triangular shell 6 v v v
Multilayered curved quadrilateral shell 8 v v v
Solid elements
Triangular prism 6 v v v
Linear hexahedron 8 v v v
Tetrahedron 4 v v v
Isoparametric triangular prism 15 v v v
Isoparametric hexahedron 20 v v v
Isoparametric tetrahedron 10 v v v
Mass elements
One node mass element 1 v v v

2.1.1 The Coordinate Systems

Wadam uses three different coordinate systems for single-body hydro models:
• A global coordinate system
• An input coordinate system
• Local finite element coordinate systems
For multi-body hydro models a more general definition applies to the coordinate systems; see Section 2.1.6.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 7

The global coordinate system, denoted (Xglo , Yglo , Zglo ), is right handed with the origin in the still water level.
The Z-axis is normal to the still water level and the positive Z-axis is pointing upwards as shown in the
Figure 2.3 and Figure 2.4. By default, the results refer to the global coordinate system. The origin is termed
the results reference point or the motion reference point. It is also possible, for the user, to specify another
result reference point.

Figure 2.3: 2D representation of a TLP hull with input coordinate system below the swl

Figure 2.4: 2D representation of a semi-submersible with centre of gravity along the global
z-axis

The input coordinate system, denoted (xinp , yinp , zinp ), is the coordinate system in which the hydro model
and the structural model are defined. Wadam imposes the restriction that all input models, i.e. the panel,
Morison, mass and FE models must be modelled with the same input coordinate system. The coordinates of
the off-body points are given in the global coordinate system in HydroD.
If the Hydrodynamic preprocessor HydroD is used, there are no constraints on the position or orientation of
the input coordinate system. The input coordinate system is specified in relation to the global coordinate
system by 3 translations and 3 Euler angles.

Local finite element coordinate systems are also used in the hydro model when this includes a Morison
model. These local coordinate systems are discussed in the sections where they are referred to.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 8

2.1.2 The Panel Model

The panel model is used to calculate the hydrodynamic loads and responses from potential theory.

The panel model may be a single superelement or a hierarchy of superelements. It may describe either the
entire wet surface or it may take advantage of either one or two planes of symmetry of the wet surface.
With symmetry planes employed the computational effort to solve the potential problem is reduced both
with respect to CPU and disk space resources. When the panel model includes more than one body, see
Section 2.1.6, there is the restriction that no planes of symmetry can be exploited.
The symmetry plane option requires that the basic part, i.e. the actually modelled part, is modelled on
the positive side of the symmetry planes as shown in Figure 2.5. Figure 2.6 also shows the basic part of a
Tension Leg Platform (TLP) panel model.

Note: No panels are allowed in the symmetry plane(s).

Figure 2.5: Symmetry plane definitions

Figure 2.6: The basic part of a TLP panel model

The basic part of a panel model consists of quadrilateral or triangular panels representing the wet surfaces
of a body. The panel model can be modelled in GeniE or Patran-Pre using standard finite elements, or within
HydroD from a section model. If more than one superelement is modelled the superelements must be
assembled in Presel. The elements accepted in the panel model are defined in Table 2.1.

Note that panels are constructed by drawing straight line segments between the corner nodes of the finite
element sides. Twisted panels are then forced to be planar by projecting element corner nodes onto panel
vertices in the plane defined by the line segment midpoints. The plane of the panel is defined by a point
being the mean of the element corner points and a normal vector being the cross product of the directional
vectors of the lines between the straight edge midpoints.

The wet surface of a panel model is identified by defining a dummy load on the panel model in GeniE or
Patran-Pre by applying a so-called HYDRO-PRESSURE load with load case number one to all external wet
surfaces of the model. Panels will be generated for all element sides below the still water level and above
the sea bed where HYDRO-PRESSURE load is defined. The model may be verified in HydroD by displaying

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 9

the mesh with dummy hydro pressure illustrated by arrows pointing from the fluid onto the wet element
sides. In Patran-Pre the corresponding option is the Hydro, Element Uniform load.

Wadam will automatically adjust the wet element sides extending above the still water level into panels with
its uppermost edge and vertices in the still water level. Depending on the shape and orientation of the wet
element sides this may actually lead to either an adjustment or a division of a wet element side into new
panels as shown by examples in Figure 2.7. This automatic algorithm is also used to adjust panels extending
below the sea-bed.

Figure 2.7: Panel adjustments at the still water level

The Wadam data check reports all the panels extracted from the input panel model noting specifically those
panels which have been adjusted or divided into two separate panels.

The modelling of a panel model with thick shell elements, which often is the case when the panel model is
defined as the wet surface of a structural model, may introduce significant deviations in the representation
of volume and area exposed to wave forces. It may be required to establish a separate panel model with
nodes on the outer surface of the thick shell structural model.

From version 9.3 of Wadam, a panel model can be defined on the free surface either in the interior domain
for removal of irregular frequencies or in the exterior domain for assigning of free surface damping. In either
of these applications, the panel elements on the free surface should be defined as a separated dummy body
in each domain with the dummy hydro pressure pointing upwards.

The following items should be remembered when modelling a panel model:

• The pressure variation as a function of draught is at a maximum close to the surface, i.e. small
element heights should be used close to the surface.

• The pressure variation is large close to the edges in the structure, i.e. element sizes normal to the
edge length should be small.

• The elements should be modelled as planar as possible. Elements which are too twisted should be
divided into smaller elements.

• Large changes in element size for neighbouring elements should be avoided.

• The largest element diagonal should be smaller than 1/6 of the smallest wave length analysed.
This requirement shall also be applied for the free surface lids described in Section 2.6.3 and Sec-
tion 2.6.4.

It is also noted that the mesh refinement of a hydro model normally needs to be high if the mean drift forces
by pressure integration shall be obtained. This compared to the calculation of rigid body responses where
reliable results may be obtained with a considerably coarser mesh.

2.1.3 The Morison Model

The Morison model is used to calculate hydrodynamic loads based on Morison’s equation. In addition to
representing the complete or parts of the structure the Morison model is used to include external forces
from mooring lines and tethers in a hydro model.

The Morison model is put together from a set of Morison elements. The Morison elements are based on 2
node beam elements and single nodes in a first level superelement generated by Patran-Pre or GeniE. The
Morison elements are actually defined by assigning hydrodynamic properties to nodes and beam elements
in HydroD.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 10

The different types of Morison elements available for calculation of hydrostatic and hydrodynamic effects
are:

• 2D Morison elements for calculation of hydrostatic and hydrodynamic loads on wet 2 node beam
elements

• 3D Morison elements for calculation of hydrostatic and hydrodynamic loads in three directions in
specific nodes

• Pressure area elements for calculation of hydrostatic and hydrodynamic loads at the ends of 2D
Morison elements

• Anchor and TLP elements for calculation of additional restoring contributions in specific nodes
Note: The Morison model must be modelled as one single first level superelement. No symme-
try plane options are available for the Morison model.
Figure 2.8 shows a beam model and Figure 2.9 the corresponding Morison model representation with 2D
Morison elements and pressure area elements. The modelling principles for establishing different types of
Morison elements are discussed in the following subsections.

Figure 2.8: 2 node beam element model of a semi-submersible platform

Figure 2.9: Morison model representation of a semi-submersible platform with 2D Morison

elements and pressure area elements

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 11

2D Morison Elements
A 2D Morison element is defined as a 2 node beam element in a preprocessor and assigned a section number
to be matched by a hydro property section specified in HydroD.

A 2D Morison element is used to include added mass and drag forces according to Morison’s equation; see
Section 2.6.6. It is also used to include hydrostatic restoring contributions.

The hydro property description for a 2D Morison element include added mass and viscous drag coefficients
in the two directions perpendicular to the longitudinal element axis. The hydrodynamic coefficients are
specified in a coordinate system local to each 2D Morison element; see Figure 2.12. In addition, the diameter
of the element is specified. The length and diameter may either be taken from the preprocessor generated
beam model or specified by HydroD.

The section numbers play an important role in the analysis of a Morison model. The use of section numbers
to include different hydrodynamic effects in the Morison model should therefore be carefully planned before
section numbers are assigned to 2D beam elements in the preprocessor. E.g. if different drag and inertia
coefficients shall be used in different locations this should be reflected by the section definitions.
The hydrodynamic coefficients specified for a 2D Morison element apply to circular cross sections. For ele-
ments with non-circular cross-sections the hydrodynamic coefficients in the ξ and ζ directions are directly
related to an equivalent cross-sectional diameter. Wadam calculates this equivalent diameter as the circum-
scribing diameter shown for the examples in Figure 2.10.

Figure 2.10: Equivalent diameters for non-circular cross sections

The 2D Morison elements may be divided into sub-elements each of which may be subjected to different
hydrodynamic forces. By default, the sub-elements have equal lengths within each element.

Figure 2.11 shows three different locations of a Morison element with respect to the still water level. The
figure illustrates how Wadam automatically performs the subdivision of elements depending on whether
they intersect the still water level or not.

Figure 2.11 (a) shows an element with five sub-elements as specified in HydroD. It has one node on each
side of the still water level. This element is first divided into five sub-elements. The sub-element intersecting
the still water level is further divided into two new sub-elements such that a sub-element border lies in the
still water level. The result is actually six sub-elements of which the first five receives hydro loads.
Figure 2.11 (b) shows elements with one node below and one in the still water level. Note that the sub-
element numbering starts at the deepest node irrespective of which is the first and which is the second
node when defining the Morison element. This forced ordering of sub-element numbers is performed for all
elements with one node below and the other either at or above the still water level. For completely sub-
merged or dry elements the sub-element division is straight forward, i.e. increasing sub-element numbers
from the first to the second node; see Figure 2.11 (c).

The local coordinate system for a 2D Morison element (ξ , η , ζ ) is defined as follows:

• The ξ -axis is normal to the element and parallel with the xy-plane of the input coordinate system.
• The η -axis points along the element from node N1 to node N2 where N1 is the first node defining the
2D Morison element as shown in Figure 2.12 (a).

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 12

 Figure 2.11: The sub-element division of 2D Morison elements

• The ζ -axis is the third axis in the right handed cartesian coordinate system defined by ξ and η .
• The ξ -axis is parallel with the xinp -axis if the η -axis is parallel with the zinp -axis; see Figure 2.12 (b).

3D Morison Elements
A 3D Morison element is defined in HydroD and can only be connected to nodes in the Morison model.

A 3D Morison element may be used to include loads which cannot be represented with a 2D Morison element
in a hydro model. Drag forces and added mass forces in the longitudinal direction of a 2D Morison element
are examples of forces that can be included with a 3D Morison element.

A 3D Morison element may be viewed as a submerged sphere which can receive both hydrostatic and
hydrodynamic loads. It will not contribute to the restoring matrix.

The hydro property description for a 3D Morison element includes added mass and viscous drag coefficients
in three directions together with a diameter of the submerged sphere.

The local coordinate system for a 3D Morison element (ξ , η , ζ ) will by default coincide with the coordinate
system of the Morison model (xinp , yinp , zinp ). If the local coordinate system shall be different from that of
the Morison model a guiding point defining the local η -axis must be specified. Figure 2.12 shows this with
node N1 being the 3D Morison element and node N2 being the guiding point. The ξ - and ζ -axes are defined
as described above for 2D Morison elements.
The forces on a 3D Morison element is acting at the node to which the 3D Morison element is connected.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 13

 Figure 2.12: Local coordinate system of 2D and 3D Morison elements

Figure 2.13: The pressure area element definition

Pressure Area Elements

A pressure area element is defined in HydroD and connected to a node in the Morison model. The node
represents the centre of the circular pressure area element. The direction of the element is defined by a
guiding point and the area with diameter d as shown in Figure 2.13.

The pressure area element includes the hydrostatic and Froude-Krylov pressure force. (The Froude-Krylov
force is the force due to the undisturbed incoming wave.) The force is applied in node N1 in the direction
of the element normal, i.e. from N1 to the guiding point. The Froude-Krylov pressure force represents a
simplified approximation to the correct pressure. Therefore 3D Morison elements must be used to include
the end effects of added mass and viscous damping.

The pressure area element has three main application areas. These are:

• Include normal pressure at submerged ends of a cylinder; see Figure 2.14 (a)
• Adjust (subtract) the pressure at intersection between cylinders; see Figure 2.14 (b)
• Include longitudinal forces due to varying cylinder diameters; see Figure 2.14 (c)
The use of pressure area elements in the composite hydro model types requires some special attention.
This is described in more detail in Section 2.1.4.

Note: Contrary to some of the other element types specified directly in HydroD the nodal loads
from point mass elements will be mapped onto the structural model.

Mooring Elements
A mooring element is defined in HydroD and can only be connected to nodes in the Morison model. Ap-
pendix A 2.1 includes a detailed description of the mooring element formulation. The mooring element is
also termed anchor element.
A mooring element may be used to include external restoring forces from weightless mooring lines with
linear stiffness characteristics.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 14

 Figure 2.14: Application areas for the pressure area element

The mooring elements are connected to nodes in the Morison model. The first connection node for the
mooring element is the guiding point, also termed the fairlead. The second connection point may be at a
windlass as shown in Figure 2.15 (a). The two mooring element connections may optionally be the same
node. The hydro properties of a mooring element include the element orientation, the pre-tension and the
restoring characteristics.

The element orientation includes two different angles; The angle αinc between the still water level and the
mooring line and the angle αx between the positive x-axis and the mooring line as shown in Figure 2.15 (b).
Note that the angle αinc ≤ π/2 with respect to the negative x-axis for nodes with x < 0.

Figure 2.15: Mooring element definitions

The restoring contributions from the mooring elements are assembled into the body restoring matrix and
hence contribute to the rigid body motion. The rigid body motion computed yields dynamic restoring forces
acting in the mooring element nodes. They are mapped onto the structural model as nodal loads. No nodal
moment loads are transferred to the structural model.

Note: Spring elements on the Input Interface File are neglected.

TLP Mooring Elements

A TLP mooring element is defined in HydroD and connected to nodes in the Morison model. It is based on the
formulation given in [5]. In addition Appendix A 2.2 summarises the description of the TLP mooring element
formulation.

A TLP mooring element may be used to include external restoring forces from a weightless tether with linear
tether characteristics.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 15

The hydro properties of a TLP mooring element include the length L of the tether, the pre-tension and the
elastic stiffness parameter λ. A horizontal offset position xoffset , yoffset may also be specified as shown in
Figure 2.16. Note that the tether length shall be the actual length at the offset position.

Figure 2.16: The TLP mooring element

The restoring contributions from the TLP mooring elements are assembled into the body restoring matrix and
hence contribute to the rigid body motion. The rigid body motion computed yields dynamic restoring forces
acting in the mooring element nodes. They are mapped onto the structural model as nodal loads.

Note: No nodal moment loads are transferred to the structural model.

2.1.4 The Composite Model

The composite model is a hydro model suitable for structures consisting of both slender and large volume
parts. The slender parts are represented with a Morison model and the large volume parts with a panel
model.

The hydrodynamic forces on a composite model are computed from potential theory for the panel model
and from Morison’s equation for the Morison model. The hydrodynamic exciting forces and matrices from
both theories are accumulated in the system of equation of motions for the composite model.

The wave kinematics applied in Morison’s equation may either be taken from the incident wave field. or it
may be specified to depend on the diffracted wave field generated from solving the diffraction problem for
the panel part of the composite model. Figure 2.17 shows a composite model where the risers, modelled
with 2D Morison elements, may optionally be exposed to loads from a diffracted wave field caused by the
shaft of the large volume structure.

Figure 2.17: Example of a composite model with a panel model and a non-overlapping Morison
model

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 16

With a composite model the pressure area element shall normally be included in the Morison model for all
wave lengths.

2.1.5 Single Super element Composite model

From version 8.1-09 of Wadam, the beams receiving loads from the Morison model and the shells or solids
receiving loads from the panel model may be modelled in the same 1. level super element. The panel model
may then be defined separately, whereas the Morison model is the same super element as the structural
model. It is also possible to have the panel model, the Morison model and the structural model all in the
same super element.

From version 9.1 of Wadam, only the active nodes and elements in the Morison model are taken into account.
The accumulation of Morison element for the 2D Morison Section only takes the sections which have been
defined as input. All other elements will be ignored. Moreover, only the nodes which are connected to the
effective Morison elements are counted. With this treatment, it is more practical to use large model as
Morison model, especially for the single super element composite model in load transfer analysis.

2.1.6 Multi-Body Modelling

The hydrodynamic and mechanical interaction between a number of structures can be analysed with the
multi-body option. The hydrodynamic interaction is computed from the potential theory as applied for a
single structure with the principal extension that the number of degrees of freedom is increased from 6
to 6N where N is the number of structures. A stiffness coupling between structures cannot be described
directly in Wadam.

Each of the bodies may be represented with a hydro model and optionally a structural model and a mass
model. The bodies may be either fixed or floating.

A hierarchical set of coordinate systems is introduced in which the individual structures and their input
models are specified. The coordinate systems applied in a multi-body analysis are therefore different from
those of a single-body analysis; see Figure 2.18. The coordinate systems are defined as follows:

• The global coordinate system (Xglo , Yglo , Zglo ) is a right handed Cartesian coordinate system with
its origin at the still water level and with the z -axis normal to the still water level and the positive
z -axis pointing upwards.
• The individual body coordinate systems (xB , yB , zB ) of each structure are specified relative to the
global coordinate system.

• The input coordinate system (xinp , yinp , zinp ) of each input model included in a body is specified
relative to the body coordinate system of that body.

Figure 2.18: The hierarchical coordinate system definition

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 17

The body independent coordinates are described in the global coordinate system, e.g. the fluid kinematics
evaluation points.

The coordinates related to a particular body are described in the corresponding body coordinate systems,
e.g. the result reference coordinate system.

The coordinates related to the individual input models are described in the input coordinate systems, e.g.
nodal coordinates of the input models.

Multibody mass modelling

The multi-body mass modelling is done by specifying the mass matrix for each body. The inertia properties
are given in the body coordinate-system.

Multibody damping
The damping between the bodies because of wave-generation (the potential damping) is computed by the
program. In addition, an external damping may be specified as input. This is given in the form of damping-
matrices for each body and for the damping between the bodies.
The damping force on body i, because of a motion of body j is then to be given as a set of damping matrices
for all pairs of bodies, including the damping of body i because of the motion of body i. Generally, these
damping matrices are defined such that they map the motion of body j in the body coordinate-system of
body j onto the force (and moment) on body i in the coordinate-system of body i.

Multibody restoring
The hydrostatic restoring of each body is computed by the program. In addition, an external restoring may
be specified as input. The external restoring is defined rhe same way as for the external damping.

Sequence of the bodies

From version 9.3, Wadam can perform analysis on a floating body with a fixed one. However, a fixed body
should be given in the order after all floating bodies. In case a dummy body for a free surface lid is used,
such dummy body should be given after all other "regular" bodies. Moreover, the dummy body for an interior
free surface lid should be given after all other type of bodies. One may check Section 2.6.3 or Section 2.6.4
for the calculation methods when a free surface lid is included as a dummy body. It should be noted that, for
a multi-body analysis with mixed floating and fixed bodies (or with free surface lids), responses in the result
interface file are only available for the floating bodies.

An analysis is regarded as a hydrodynamic multi-body analysis if more than one regular body is defined.

2.1.7 Mass Modelling

Global mass information is required in Wadam for analysis of floating structures. The mass is used both in
the hydrostatic calculations to report imbalances between weight and buoyancy of the structure and in the
equation of motion.

Wadam provides two methods to establish global mass matrices:

• Direct input specification of a global mass

• Assembling of a global mass matrix from a mass file (no utilisation of symmetry planes)
Wadam transfers accelerations to Loads Interface Files for subsequent structural analysis. A consistent
calculation of inertia loads from these accelerations is ensured in Wadam and Sestra by access to the same
module for finite element mass generation.
For beam element models there is the alternative option to calculating inertia loads in Wadam and transfer-
ring the loads to the Loads Interface Files.

The remaining part of this section describes the two methods for establishing global mass information. It
should be noted that for either of the two methods, the mass matrix is by default calculated by HydroD, from
the specified user input.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 18

Direct Input Specification of Global Mass
The direct input specification of a global mass comprises giving the total mass of the structure together
with the centre of gravity, the gyration radii and the products of inertia. The centre of gravity is specified
with respect to the input coordinate system. The gyration radii and the products of inertia are specified with
respect to the global coordinate system. In HydroD there are more options to choose the coordinate system
from which the user specified mass is given.

Alternatively, a 6 by 6 mass matrix can be specified as direct input to Wadam from HydroD.

The direct input specification of a global mass cannot be used together with the option to integrate forces on
sectional planes in the hydro model. This is because the global mass matrix does not include any information
of the mass distribution on the particular elements and hence the inertia force contributions from these
elements cannot be computed. As described below the option to generate the mass from a mass model
must be used to obtain these sectional forces.

Assembling of Global Mass Matrix from Mass Model

The mass model used in the assembling of a global mass matrix may be defined by using a finite element
model built from an arbitrary superelement hierarchy. With this option mass contributions are assembled
from the element types defined in Table 2.1.

The assembling of a global mass matrix from a mass model should be used together with the option to
integrate forces on sectional planes in the hydro model.

The mass model must be defined in the same coordinate system as used for the other input models. This is
described in more detail in Section 2.1.1.

The mass model may be identical to the structural model or it may be a completely different superelement
hierarchy.

Alternatively, a point mass file, which describes a list of point mass and its coordinates, can be given as
mass model.

2.1.8 Structural Modelling

Wadam may be used to calculate hydrostatic and hydrodynamic loads on a structural model; see also Sec-
tion 2.3.

The structural model may be built from an arbitrary large superelement hierarchy. It may include any of the
finite element types defined in Sesam. However, only the finite element types listed in Table 2.1 will receive
hydrostatic and hydrodynamic loads.

The finite-elements which shall receive hydrostatic and hydrodynamic loads must be identified in the mod-
elling phase. The technique to identify elements differs between beam elements and element types with
surfaces (shells and volumes) as described in the two subsections below.

Nodal accelerations from rigid body motion will be calculated for all the nodes in a structural model. If
the structural model consists of more than one superelement a combination of loads from different su-
perelements is required in the super element assembling performed in Presel. The combination of loads is
described in Section 2.1.10.

Loads on Super elements with Beams

The beam elements and nodes which shall receive hydrostatic and hydrodynamic loads from Wadam must
be included in a Morison model. The loads are transferred to the beam elements and nodes which are
connected to Morison elements in HydroD: see Section 2.3.5.

Note that both hydrostatic and hydrodynamic load contributions on free ends of beam elements require the
element ends to be closed with Morison pressure area elements as described in Section 2.1.3.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 19

Loads on Super elements with Shell and Solid Elements
For shell and solid elements loads are transferred to the finite element sides which are identified as wet. In
GeniE this is done by assigning a wet surface property to the wet plates. The dummy hydro pressure load
defined in GeniE (or Hydro, Element uniform load in Patran-Pre) is used to identify the wet element sides.
The definition of wet sides of the structural model is equivalent to the definition of panels in the panel model;
see Section 2.1.2. Wet element sides may be included in several superelements of a structural model.

Wadam transfers pressure loads to both the external wet surface of a structural model and to the wet
surfaces of internal tanks. The dummy load case number of the HYDRO-PRESSURE load must be used to
identify which of the wet elements shall receive external and which shall receive internal HYDRO-PRESSURE
loads. The rules for this dummy load case numbering is the following:

External wet surface: The HYDRO-PRESSURE load case number must be equal to one.

Internal wet surface: The HYDRO-PRESSURE load case for the first internal wet surface (tank) must be
equal to two. Additional internal tanks are numbered consecutively with load case
number three assigned to tank number two and so on.

It is important in the definition of wet element sides that the direction of the pressure load is pointing from
the fluid towards the element side. For this purpose Patran-Pre or GeniE provides the option to visualize
the direction of the pressures defined on the finite-element mesh. Figure 2.19 shows an idealised view of
the normal vectors pointing towards wet element sides. This load can also be visualized and verified in
HydroD.

Figure 2.19: Idealised view of hydro pressures on structural element sides

The dummy load cases used to identify wet structural elements will not be in conflict with load cases gener-
ated by Wadam or other load cases defined by the pre-processors.

2.1.9 Free Surface Modelling

The free surface model used in the second order sum- and difference- frequency force calculation in Wadam
may be generated by GeniE, Patran-Pre or HydroMesh. It may optionally also be interpreted directly from
the Wamit version 5.3S free surface format, ref. [2].
The part of the free surface actually modelled by surface panels is defined by the radius of a circle as shown
in Figure 2.20. This so-called partitioning radius R must enclose the hydro model. It should be determined
according to the decaying rate of local waves. An appropriate approximation is R ∼ O (h) for shallow water
and R ∼ O (λ) for deep water. Here h is the water depth and λ  h is the longest wave length involved. The
ratio h/λ may have to be substantially larger than 1 to achieve accuracy in deep water, ref. [3].

The free surface must be meshed with 4 node shell elements (no 3 node elements). The HYDRO-PRESSURE
load case can point in either positive or negative z-direction.

The free surface model shall have the same symmetry properties as the panel model.

2.1.10 Load Case Numbering and Load Case Combinations

The default load case numbers LC generated by Wadam on the Loads Interface Files are given from the
following equation:

LC = (((MM − 1) · NOK + (LL − 1)) · NPHA + IPHA) · I · NSEL + ISEL (2.1)

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 20

 Figure 2.20: Free surface mesh

where

MM is the actual heading number

NOK is the total number of frequencies

LL is the actual frequency number

NPHA is the number of phase angles if the time output option is specified, NPHA = 1 if
complex loads are generated

IPHA is the actual phase angle number, IPHA = 1 if complex loads are generated

I is zero for static load cases and one for dynamic load cases

NSEL is the number of occurrences of the first level superelement in question

ISEL is the actual occurrence number of the first level superelement in question

For a given first level superelement with complex loads Equation (2.1) will generate load case numbers 1
through NSEL as static load cases and load cases number NSEL + 1 through NOK · NOH · (NSEL + 1) as
dynamic load cases. Table 2.2 illustrates the correspondence between superelement occurrence and wave
frequency and heading angle.

The relation between superelement occurrence numbers in Wadam and the superelement index numbers in
Presel is important when performing load case combinations in Presel. This is discussed in Example 2.2.
For loads transferred to a structure modelled with shell or solid elements Wadam includes some options
to manipulate the generation of loads on the Loads Interface Files and the numbering of load cases. More
specifically, for floating structures Wadam by default generates four different types of loads represented as
static and dynamic loads respectively. These are:

• Hydrostatic pressure and gravity summed together in the first global load case.
• Hydrodynamic pressure loads and nodal accelerations summed together for each combination of
incident wave frequency and heading angle into global load cases starting with load case number
two.

For the case of load transfer from a panel model to a shell/solid model each of the four load types above
may optionally be suppressed. This is controlled from HydroD.

For structural models consisting of one superelement only Wadam will by default generate a hydrostatic
load case for the still water condition and a sequence of hydrodynamic load cases, one for each specified
combination of wave frequency and wave direction. If deterministic load calculation is specified separate
load cases are also created for each specified phase angle. The load cases generated by Wadam will be the
global load cases for single superelement structural models.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 21

For structural models defined from a hierarchy of superelements the hydrostatic and hydrodynamic load
cases for each superelement will include the load cases for all the occurrences of each particular superele-
ment. Equation (2.1) is used to assign the unique load case numbers for all occurrences of superele-
ments.
Furthermore, for structural models defined from a hierarchy of superelements the load cases created by
Wadam for first level superelements are combined into new higher level load cases in Presel. The load
case combination is performed recursively, i.e. repeated for each new higher level superelement created
in Presel. At the structure level the load case combinations coincides with the global load cases. See
also the Presel User’s Manual for a description of the assembling of superelement hierarchies and load
combinations.

The term superelement occurrence number defines the actual location of a superelement in a superele-
ment hierarchy. The number is found by counting the occurrences of a superelement from left to right in a
superelement hierarchy (or from top and down in hierarchy as printed by Presel).

Two simple examples are included to describe the relation between global load case numbers and the load
case numbers generated by Wadam. In both examples the set of global loads include one static load case
and 6 dynamic load cases defined as the combinations of three incident wave frequencies and two heading
angles.
Example 2.1: Load case numbering for a single superelement model

This example consists of a structural model built from one single superelement. Here the load case numbers
generated by Wadam directly coincides with the global load case numbers. Table 2.2 shows the correspon-
dence between the global load case numbers and the wave frequency and heading combinations.

Table 2.2: Load case numbering for a single superelement structural models

Load case number

Load case description
(global = Wadam)
1 Hydrostatic
2 β = 0◦ , ω = 0.1
3 β = 0◦ , ω = 0.2
4 β = 0◦ , ω = 0.3
5 β = 90◦ , ω = 0.1
6 β = 90◦ , ω = 0.2
7 β = 90◦ , ω = 0.3

Example 2.2: Load case numbering for a model assembled in a superelement hierarchy

This example is a model consisting of a first level superelement, superelement number 10, used in two
different positions in a two level superelement hierarchy. The top level superelement number is 100. Adopt-
ing the terms used above there are two occurrences of the same first level superelement in this hierarchy.
Figure 2.21 shows the superelement hierarchy.

Figure 2.21: Two-level superelement hierarchy with occurrence 1 and 2 of superelement 10

included in superelement 100

Table 2.3 shows the correspondence between the global and Wadam generated load case numbers. The
table also includes the superelement occurrence numbers, a description of the separate load cases and the
Presel generated superelement index numbers.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 22

 Table 2.3: Load case numbering for a model assembled in a superelement hierarchy

Global load Wadam load Load case Occurrence of Presel

case number case number description superelement 10 index
1 1 static 1 1
1 2 static 2 2
2 3 β = 0◦ , ω = 0.1 1 1
2 4 β = 0◦ , ω = 0.1 2 2
3 5 β = 0◦ , ω = 0.2 1 1
3 6 β = 0◦ , ω = 0.2 2 2
4 7 β = 0◦ , ω = 0.3 1 1
4 8 β = 0◦ , ω = 0.3 2 2
5 9 β = 90◦ , ω = 0.1 1 1
5 10 β = 90◦ , ω = 0.1 2 2
6 11 β = 90◦ , ω = 0.2 1 1
6 12 β = 90◦ , ω = 0.2 2 2
7 13 β = 90◦ , ω = 0.3 1 1
7 14 β = 90◦ , ω = 0.3 2 2

In this very simple example there is a one-to-one correspondence between occurrence numbers and Presel
index numbers. This one-to-one correspondence is not required. However, when the occurrence numbers
and the Presel indices do not match the load combination in Presel should be carefully performed to ensure
that the indices correspond to the correct occurrence numbers.

The one-to-one correspondence may be violated for particular choices of using the same first level superele-
ment in several sub-hierarchies of a multi level superelement model.

2.2 Global Response Analysis

2.2.1 General

The global response feature in Wadam computes the response of fixed and floating structures due to wave
loads. Results computed are forces and response transfer functions assuming rigid bodies. Also global
matrix results data, sectional loads and off-body kinematics results may be produced. The results can be
transferred to the statistical post-processor Postresp for graphics presentation and further results processing
through a Results Interface File.

The statistical postprocessing in Postresp consists of statistical analysis of transfer functions including cal-
culation of response spectra and short & long term statistics. Postresp also includes the option to calculate
the equation of motion from the global matrices and exciting forces transferred from Wadam.

The global response analysis is performed for a system consisting of a hydro model and a mass model.
The latter is only required if the system is specified to be floating. Additional mooring and tether stiffness
characteristics may also be provided for floating systems.

The hydro model may consist of either a panel model, a Morison model built from beam elements or a
combination of these two model types. The hydro model represents different types of hydrostatic and
hydrodynamic loads. The hydro model concept is described in detail in Section 2.1.
A mass model is required when the global response analysis includes calculation of motions. A mass model
may either be defined as a FE model where the mass of each finite element contributes to a 6 by 6 body
mass matrix, also termed the global mass matrix. Alternatively the mass model may be specified directly
as an input global mass matrix. The mass model is described in more detail in Section 2.1.7.

2.2.2 Computation of Wave Loads

Wadam calculates wave induced forces for a specified set of wave frequencies and heading angles by one
of the following three calculation methods:

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 23

 • Morison’s equation applied to a Morison model
• The MacCamy-Fuchs formula applied to a Morison model
• The potential theory applied to a panel model
• A method in which Morison’s equation and the potential theory both are applied to compute hydro-
dynamic loads on the same hydro model. This type of hydro model is termed a composite model
and is described in Section 2.1.4. With the composite model the wave kinematics applied in Mori-
son’s equation may optionally be modified to take into account the diffracted wave field due to the
presence of a panel part in the hydro model.

The motion responses for a hydro model is obtained by solving the equations of motion for a set of wave
frequencies and heading angles. The rigid body added mass, damping and restoring matrices used in the
equations of motion may be calculated by applying Morison’s equation, the potential theory or the composite
method as described above. Except for frequency dependent added mass matrices these matrices may
alternatively be specified directly by the user.

2.2.3 The Global Response Results

The wave induced forces and moments and the motion responses calculated by Wadam are reported with
respect to a motion reference point which is, by default, located at the intersection between the still water
level and a vertical line through the common origin of the models used in the analysis. The coordinate
systems in Wadam are described in Section 2.1.1.

The results available from a global response analysis of a hydro model include transfer functions for:
• Wave exciting forces and moments
• Motion responses
• Sectional loads
• Rigid body matrices
• Off-body kinematics
• Surface elevations
Section 2.5 describes all these result types in more detail.

2.3 The Calculation of Detailed Loads on a Structural Model

2.3.1 General

The detailed load calculation feature provides a tool for automatically transferring wave loads from a hydro-
dynamic analysis into finite element loads for a structural analysis.
Both hydrostatic and hydrodynamic loads on a structural model may be calculated. The results, which may
be both FE pressure loads and nodal loads, will be transferred to Loads Interface Files for subsequent linear
static or dynamic analysis in Sestra. Wadam also transfers environmental information and load case num-
bering information to Sestra on a separate file (the S-file). The latter is specifically required for a subsequent
stochastic fatigue analysis using Framework or Stofat.

2.3.2 The Structural Load Types

Three different load types may be generated by Wadam and transferred to a structural model. These
are:

• Hydrostatic loads with contributions from forces in the still water condition and pre-tension from
mooring and tether systems

• Gravity load representing the weight of the structure

• Hydrodynamic loads with contributions from exciting forces from incident waves, forces from wave
induced motion and rigid body accelerations

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 24

Detailed descriptions of distributed loads are included in Section 2.5.15 and Section 2.5.16.

Wadam generates separate load case numbers for the hydrostatic load and for each of the hydrodynamic
loads, i.e. for each wave frequency and heading. When having a superelement hierarchy model, these
loads may be combined into new load cases with Presel. The new combined load cases will be used during
subsequent structural analysis and postprocessing. Section 2.1.10 contains a more detailed description of
load case numbering and load case combinations in Sesam.

2.3.3 Deterministic Loads

Wadam also provides the option to report transfer functions for FE loads as deterministic loads (time do-
main). That is loads represented for specified phase angles of incident waves with given wave ampli-
tudes.
The deterministic results presentation may also be used together with the option to calculate the following
types of loads:
• Non-linear viscous drag forces from Morison’s equation for fixed structures.
• Pressure loads up to the instantaneous free surface, see Section 2.6.9.
• Time invariant current profiles added to the incident wave field in the calculation of forces by Mori-
son’s equation. See Section 2.4.2 for the description of current profiles in Wadam.

2.3.4 Detailed Loads Transfer to a Model with Shell or Solid Elements

The transfer of detailed loads to a large volume structure modelled with shell or solid elements is performed
as follows:

• The hydrostatic load at still water condition is calculated as normal pressures directly on the shell
and solid element surfaces defined as wet sides.

• The hydrodynamic pressure loads on shell or solid elements can only be obtained from a potential
theory calculation based on a panel model. The actual mapping of panel pressures into normal
pressure loads in a structural model is automatically performed by the algorithm described in Ap-
pendix A 3.3.

• Independent nodal loads require a Morison model to be used together with the panel model. More
specifically, if loads from mooring lines and tethers shall be included in a structural system a Morison
model consisting of the connection nodes must be created.

2.3.5 Detailed Loads Transfer to a Model with Beam Elements

The transfer of loads to a slender structure, for example a jacket or a jackup, modelled with beam elements
is performed as follows:

• The hydrostatic load at the still water condition is represented as loads on beam elements and nodes
in the structural model.

• The hydrodynamic loads are calculated for the beam elements and nodes corresponding to the Mori-
son elements. The hydrodynamic pressure loads obtained by Morison’s equation are represented
directly as line loads and nodal loads on the beam element model to be transfered to structural
analysis.

2.4 Environmental Description

2.4.1 Surface Waves

The models in Wadam may when first order potential theory and Morison’s equation are applied be exposed
to planar and linear harmonic waves, i.e. waves described by the Airy wave theory. For the second order
option for the potential theory see [3] for a detailed description of the theoretical background.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 25

 Figure 2.22: Surface wave definitions

The incident waves may be specified by either wave lengths, wave angular frequencies or wave periods. The
direction of the incident waves are specified by the angle β between the positive x-axis and the propagating
direction as shown in Figure 2.22 (a).

The incident wave used in Wadam is defined as

h i
η = Re A ei(ωt−k(x cos β+y sin β)) (2.2)

which alternatively may be written as

η = A cos (ωt − k (x cos β + y sin β)) (2.3)

This represents a wave with its crest at the origin for t = 0 as shown in Figure 2.22 (b).
The fluid velocity v = vx i + vy j + vz k and acceleration a = ax i + ay j + az k for the incident waves are:

k cosh (kz + kd)

vh = vx i + vy j = Aω cos (ωt − k · x)
k sinh (kd)
(2.4)
sinh (kz − kd)
vz = −Aω sin (ωt − k · x)
sinh (kd)

k cosh (kz + kd)

ah = −ax i + ay j = Aω 2 sin (ωt − k · x) (2.5)
k sinh (kd)
sinh (kz − kd)
az = −Aω 2 cos (ωt − k · x) (2.6)
sinh (kd)

where

d Depth

k Absolute value of wave number

ω Wave angular frequency

A Wave amplitude
x = xi + yj – location in the x-y plane
k = k (i cos β + j sin β) – two dimensional wave number
z Vertical coordinate with z-axis upward, z = 0.0 at still water level

β Direction of wave propagation

The finite depth dispersion relation used in the above expressions is

ω 2 = gk tanh (kd) (2.7)

The wave period is given by

2π
T = (2.8)
ω

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 26

and the wave length is

2π
λ= (2.9)
k

For a more detailed description of linear wave theory see [8].

The fluid kinematics above the still water level is obtained by constant extrapolation in Wadam.

2.4.2 Current Profiles

Wadam provides the option to specify time invariant current profiles for the calculation of deterministic
Morison element forces.

Note: The current profile can only be used for fixed structures.

The current profile may be specified at a set of positions along the z-axis of the global coordinate system.
Current values at intermediate z-positions are obtained by linear interpolation. The direction of the current
in the horizontal plane is specified at each positions.

Figure 2.23: Current profile definition

2.4.3 Water Depth

Wadam provides the option to specify a water depth. The water depth is used in two different calculation
phases in the program:
• It is used in the processing of the panel model to remove all panels below the sea-bed.
• It is used in the calculation of Green’s functions for finite water depth.

2.5 Results Types Reported from Wadam

2.5.1 Units

When performing an analysis with Sesam the user must apply a set of consistent units. The same units must
be used in all programs throughout the analysis from modelling to results presentation.

The basis for determining a set of consistent units is the fundamental equation:

f = ma (2.10)

In terms of the fundamental units of mass [M], length [L] and time [T] this equation may be written:
 
ML
F= (2.11)
T2

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 27

Force, stress, density, etc. are not fundamental units and must be derived in terms of the units of mass,
length and time. Whenever possible it is simplest to use S.I. units:

[L] Length in meters (m)

[ M] Mass in kilograms (kg)

[T] Time in seconds (s)

Force will then be in Newton (N):

kg m
1N = 1 (2.12)
s2

The units used in Wadam is controlled by the acceleration of gravity [L/T2 ] and the fluid density [M/L3 ]. All
other input data must be expressed in terms of these units. For example:

• Fluid kinematic viscosity: [L2 /T]

• Fluid velocity: [L/T]

2.5.2 Result Reference Point

For single-body structures the results from Wadam are reported with respect to a default result reference
point which is coinciding with the origin of the global coordinate system; see Section 2.1.1. For multi-body
structures there is one result reference point for each body coinciding with the body coordinate systems;
see Section 2.1.6.

The result reference point can also be specified by the user. In this case, the user specifies the displacement
of the result reference point relative to the default point. The displacement is given for all three directions of
the axes of the global coordinate-system. The results relative to the user-specified result point are obtained
by linear transformations of the end-results. The drift-forces are then not transformed correctly to second
order in the amplitude and should normally not be used in this context.

2.5.3 Dimensioning of Results

Wadam reports results on dimensionalised and non-dimensionalised form as follows:

• The Hydrodynamic Results Interface File contains dimensionalised results.

• The Loads Interface Files contain dimensionalised results.
• The Wadam print file contains both forms as follows:
– Chapters 2 and 5 contain dimensionalised results.

– Chapter 4 contains non-dimensionalised results.

Table 2.4: Dimensionalising factors for matrices

Dn entryi,j
i = 1-3, j = 4-6
Result type i = 1-3, j = 1-3 i = 4-6, j = 1-3 i = 4-6, j = 4-6
Inertia matrix ρV ρV L ρV L2
Added mass matrix ρV ρV L ρV L2
p √ √
Damping matrix ρV g/L ρV gL ρV L gL
Restoring matrix ρV g/L ρV g ρV gL

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 28

 Table 2.5: Dimensionalising factors for results

Dn modei
Result type i=1−3 j =4−6
Exciting forces ρV gA/L ρV gA
Motion A A/L
2
Drift forces ρgLA ρgL2 A2
Sectional loads ρV gA/L ρV GA
Fluid pressures ρgA
p
Fluid velocities g/LA
Fluid accelerations gA/L
2nd order forces ρgLA2 ρgL2 A2
2nd order motions A2 /L A2 /L2
2nd order pressures ρgA2 /L

The non-dimensionalising factor Dn specified in Table 2.4 and Table 2.5 may be used to obtain dimension-
alised results Fd from the formula:

Fd = Dn Fn (2.13)

where Fn is a non-dimensionalised result reported in chapter 4 in the Wadam print file. The factors used in
the tables are:

ρ Density of the fluid

g Acceleration of gravity
L Characteristic length

V Displaced volume of the body

A Amplitude of the incoming wave, equal to one for harmonic results

2.5.4 Transfer Functions and Phase Definitions

The transfer functions in Wadam describe responses for bodies in harmonic waves. The reported responses
are normalised with respect to the incident wave amplitudes. With a transfer function H(ω, β) the corre-
sponding time dependent response variable R(ω, β, t) can be expressed as

R(ω, β, t) = A Re H((ω, β)) eiωt

 
(2.14)

where A is the amplitude of the incoming wave, ω is the frequency of the incoming wave, β describes the
direction of the incoming wave and t denotes time. The phase angle φ between the incident wave and the
time varying response is defined from
h i
R(ω, β, t) = A Re |H((ω, β))| ei(ωt+φ) (2.15)

where |H| is the amplitude of the transfer function. The transfer function and the phase angle may be
expressed as
HIm
H = HRe + iHIm and φ = atan (2.16)
HRe

The time varying response can alternatively be expressed as

R(ω, β, t) = A [HRe cos ωt − HIm sin ωt] (2.17)

The phase lead φ of the response relative to an incident wave with the wave crest at the origin of the global
coordinate system is shown in Figure 2.24.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 29

 Figure 2.24: Definition of phase between the response and the incident wave

2.5.5 Hydrostatic Results

Wadam calculates the hydrostatic restoring results from the hydro model. It is given with dimensions and
include:
• The sum of displaced volume of the panel and Morison part of the model
The total displaced volume is reported together with the separate contributions from the panel and
Morison parts.

For the panel model the volume is reported from three different calculations, i.e. from summing
up of control volumes in the three different directions. The reported total volume is taken as the
median of the three volumes (not mean but middle value of three values).

Note: If the separate control volumes are differing by a significant number this normally indi-
cates that the wet surface of the model is not properly defined.

• The centre of buoyancy

• The water plane area
• The metacentric height
• The global hydrostatic restoring matrix assembled from both hull restoring and additional stiffness
terms
The additional hydrostatic stiffness terms included in the hydro model may include contributions from:

• Risers, mooring lines or tethers represented with Morison mooring or tether elements
• Stiffness matrices specified directly on the analysis control data
For multi-body models the hydrostatic restoring matrix for each body is reported.
Note: In the hydrostatic restoring contribution from the Morison elements the local waterplane
moment for each Morison member is not included. This can cause errors if some of
the beams have a large diameter. The fix to this problem is to include this restoring
contribution as an additional restoring matrix.

2.5.6 Global Mass Matrix

A 6 by 6 mass inertia coefficient matrix is reported for each body. It is generated according to the input
definition and hence may be calculated from either:

• The hydro model

• The structural model
• A specific mass model

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 30

 • The input definition of an inertia mass matrix based on:
– Specifying the centre of gravity and the radii of gyration together with the total mass

– Specifying the 6 by 6 global mass matrix

2.5.7 Added Mass Matrix

The 6 by 6 added mass matrix is reported for each separate body. The added mass interaction matrices
between any two bodies in a multi-body system are also reported. The added mass matrix is calculated
according to the type of the hydro model as follows:

• Frequency dependent added mass from potential theory

• Frequency independent added mass from Morison’s equation
For the composite hydro model the added mass at a given frequency is reported with both frequency de-
pendent and independent contributions.

2.5.8 Damping Matrix

The 6 by 6 damping matrix is reported for each separate body. The potential damping interaction matri-
ces between any two bodies in a multi-body system are also reported. The damping matrix is calculated
according to the type of the hydro model as follows:

• Frequency dependent damping from potential theory

• Frequency independent linearised viscous damping from Morison’s equation
• HydroD specified frequency dependent or frequency independent 6 by 6 damping matrices
For the composite hydro model type the damping for a given frequency is reported as both frequency de-
pendent and independent contributions.

2.5.9 Exciting Forces and Moments

The transfer functions for exciting forces and moments due to the incident waves are reported for each body
at all the combinations of frequencies and wave headings. The transfer functions for rigid body forces and
moments are calculated as follows:

• By integration of exciting forces from all types of Morison elements obtained from the linearised
Morison’s equation

• By integration of pressures on all the panels obtained by solving the diffraction problem
• By applying the Haskind relation on the radiation potentials if no detailed panel pressures shall be
calculated

For the composite model type the exciting forces from the panel and Morison parts are reported both sepa-
rately and as the combination used in the subsequent analysis, e.g. in the equation of motion.

By selecting time domain output the above results will be reported as deterministic forces and moments at
specified phases of the incident waves with given wave amplitudes.

For fixed structures the deterministic Morison option may be used to include the following non-linear effects
in Morison’s equation:

• Non-linear viscous drag formulation

• Time invariant current may be superimposed on the fluid velocities.
• Forces and moments may be calculated up to the free surface by constant extrapolation of the linear
wave profile.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 31

2.5.10 Rigid Body Motion

The transfer functions for rigid body motion due to the incident waves are reported for each body for all
the combinations of wave frequencies and heading angles. The roll, pitch and yaw motions are reported in
radians.
The equation of motion is assembled from the calculated global matrices and transfer functions as described
in the previous sections. The HydroD specified damping and stiffness matrices will be added to the otherwise
calculated matrices.
By specifying the time domain output format the motions will be reported as deterministic motions at spec-
ified phases of the incident waves with given wave amplitudes.

2.5.11 Second Order Mean Drift Forces

The second order mean drift forces due to the linear incident waves are reported both on the print file and
in the Hydrodynamic Results Interface File. They are calculated by one of the following methods:
• Momentum conservation in the three horizontal degrees of freedom
• Pressure integration in all six degrees of freedom. This method will also give the mean drift forces
on each individual body in a multi-body analysis.

2.5.12 Second Order Sum and Difference Frequency Results

The transfer functions for sum and difference frequencies are reported in the print file and on the Hydrody-
namic Results Interface File. The following second order results are available:

• The quadratic second order force

• The second order forces on the body by an indirect calculation method
• The second order pressure distribution on the body (available in the print file and in the Wamit
output file format)

• The second order wave elevation at specified points (only available in the Wamit output file format)

2.5.13 Fluid Kinematics

The transfer functions for pressure and particle velocity in specified points in the fluid is reported both on
the print file and in the Hydrodynamic Results Interface File. The fluid kinematics points are specified in the
fluid domain outside the hydro model. The fluid kinematics depends on the type of hydro model:
• It is obtained from the incident undisturbed wave field if only Morison’s equation is applied.
• It is obtained from the diffracted wave field if the potential theory is applied to a fixed structure.
• It is obtained from the radiated and diffracted wave field if the potential theory is applied to a floating
structure.

By specifying the time domain output format the fluid kinematics will be reported as deterministic pressure
and velocities at specified phases of the incident waves with given wave amplitudes.

The surface elevation η in a radiated and diffracted wave field is obtained from potential theory as
p
η= (2.18)
ρg

where p is the non-dimensionalised pressure at the still water level.

2.5.14 Wave Drift Damping

The Wave Drift Damping describes the rate of change of the Mean Drift force with forward speed computed
at zero speed. The sign of this rate of change is in most cases negative, meaning that this will represent
a damping mechanism for the slow drift motion excited by the second-order difference frequency forces or
due to the interaction of the waves with a current.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 32

The 3x3 Wave Drift Damping matrix (surge, sway, yaw) is reported in the Hydrodynamic Results Interface
File. This matrix is computed according to the methods described in [9].

The computation of the wave drift damping requires a free surface mesh which is defined as input exactly like
the free surface mesh for second-order analysis. The mesh used in the Wave Drift Damping computations
does, however, not need to have a circular outer boundary.

A Morison model can not be used if Wave Drift Damping is to be computed.

The iterative equation-solver for the potential-solution can not be applied if Wave Drift Damping is com-
puted.

2.5.15 Distributed Hydrostatic Loads

The hydrostatic loads on a structural model is calculated directly on the individual finite elements of the
model. It is represented on the Loads Interface Files as a real load case.

Structural Beam Models

For wet beam structural superelements the hydrostatic load case is represented as line loads on the beam
finite elements and as nodal loads on nodes receiving loads from 3D Morison elements and pressure area
elements. In addition, the pre-tension from Morison anchor and TLP elements are included as nodal loads in
the hydrostatic load case.
A general description of line loads on beam finite elements is included in Appendix A 3.2.

Note: Pressure Area Elements should always be defined at the end of vertical members such
as the legs on a semi-submersible.

Structural Shell or Solid Models

For wet surfaces of shell and solid elements in a structural model the hydrostatic loads are represented
as normal pressures. Because the hydrostatic pressure intensities are evaluated individually at the z-
coordinate of each node the normal pressure on wet element sides will have a variation in the z-direction as
opposed to the hydrodynamic loads which are constant over each element.

The gravity component of the static load is included by writing the acceleration of gravity to the Loads
Interface File.

2.5.16 Distributed Hydrodynamic Loads

The distributed hydrodynamic loads are calculated in a body fixed coordinate system and include the follow-
ing contributions:

• Exciting forces from incident waves

• Forces from wave induced motion
• Fluctuating hydrostatic pressure forces due to the body motion
The distributed forces presented in the Wadam print file are:

• Transfer functions for pressures on the panels in the hydro model 1

• Transfer functions for forces acting at the centre of gravity for 2D Morison elements
• Transfer functions for nodal forces acting on nodes in the hydro model which are connected to the
following Morison element types:

– 3D Morison elements

– Pressure area elements

– Mooring and TLP elements

The distributed loads transferred to the Loads Interface Files are:

1
Not including fluctuating hydrostatic pressure.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 33

 • Transfer functions for normal pressures on the wet sides of shell and solid elements. The pressures
are constant over the elements. These loads originate from pressures at the centroid of panels ac-
cording to the mapping algorithm described in Appendix A 3.3. They include fluctuating hydrostatic
pressure.
• Transfer functions for line loads on beam elements. These loads originate from forces at the centre
of gravity of 2D Morison sub-elements and dry Morison elements. The calculation of line loads on 2
node beam elements is described in Appendix A 3.2.
• Transfer functions for loads at nodes which are connected to Morison elements
• An acceleration field acting on all the nodes in the structural model
• The inertia components are only included as loads if the mass model is specified as a distributed
mass model.

Wadam and the structural analysis program Sestra both employ the standard Sesam finite element library
to generate a mass representation. This implies that the connection between accelerations and inertia loads
is consistently handled in the two programs.

2.5.17 Load Sum Reports

Load sums are important tools to verify both the correctness of input models to Wadam and of the consis-
tency between Wadam and subsequent structural analyses.

Wadam reports load sums on both the structural model and the hydro model in chapter 5.1 in the print file.
Note that these load sums are reported with dimensions.

• Load sums for the loads transferred to the structural beam model:
– The sum of hydrostatic loads transferred to a beam model is reported in the print file as
static loads. It includes the hydrostatic forces from each Morison type of element as follows:

∗ The buoyancy and pre-tension components from mooring and TLP elements are al-
ways included.

∗ The gravity components are only included if the mass model is specified as a dis-
tributed mass model.
– The sum of hydrodynamic loads acting on the beam model is reported in the print file for
each incident wave frequency and wave heading. It includes the hydrodynamic forces from
each Morison type of element as follows:

∗ The hydrodynamic pressure components acting on beams and nodes

∗ The inertia components are only included if the mass model is specified as a dis-
tributed mass model.

• Load sums for the loads transferred to the structural shell/solid model:
– The sum of hydrodynamic pressure loads transferred to the shell and solid structural model
is reported in the print file for each incident wave frequency and wave heading. It includes
the part of the hydrodynamic loads actually transferred to normal pressure loads on shell
and solid elements. Note that this load sum does not include inertia loads.

– The sum of pressures and inertia loads acting on the structural model may be calculated
from the Sestra data check where the accelerations transferred from Wadam are converted
into inertia loads; see Section 2.5.16.
• Load sums for the hydro model:
– The sum of hydrodynamic forces acting on a panel model is reported in the print file for
each incident wave frequency and heading. It includes the pressures acting at the centroid
of each panel in the hydro model. Note that this load sum does not include inertia loads.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 34

2.5.18 Sectional Loads

Wadam calculates sectional loads by integration of distributed forces on specified sides of given planes inter-
secting a hydro model. The frequency dependent exciting and inertia forces are included in the integration.
The sectional planes are specified in the input coordinate system and must be normal to one of the main
axes of the global coordinate system. A moment reference point is specified for each plane. Figure 2.25
displays a simple submerged beam model with a sectional plane xinp = 0. The moment reference point is
defined in the input coordinate system (xinp , yinp , zinp ).

Figure 2.25: Sectional loads at a plane through x = 0

The sectional loads are reported in the Wadam print file and on the Hydrodynamic Results Interface File as
complex transfer functions.

Still water sectional loads are reported on the Wadam print file.

The sectional loads are integrated with respect to a body fixed coordinate system. The sectional load on a
cut comes from the integration of F-ma over that part of the structure which is on the positive or negative
side of the cut. The components of the load are in the global system. This means that for the dynamic
loads the load computed by integration over the positive side and the load computed by integration over
the negative side will be equal in magnitude, but have a phase difference of 180 degrees.

As an example we consider a ship cut at x = 0 normal to the x-axis:

If integration over the positive side is specified the sectional load will be the force acting on the part of the
hull with x < 0 from the part with x > 0. This means that the static bouyancy force will give a positive
vertical shear force and a negative vertical bending moment. We will also find that the phase angle of the
vertical bending moment in a hogging situation is close to zero for long waves.

If integration over the negative side is specified the static bouyance force will give a positive vertical shear
force and a positive vertical bending moment. The phase angle for the vertical bending moment in a long
wave will be close to +/- 180 degrees.

The contributions to the sectional loads from different hydro models are computed as follows:

• For Morison models the exciting and inertia forces, obtained at the centre of gravity of each Morison
sub-element on the specified side of the sectional plane, are included.

• For panel models the exciting forces at the centroid of each panel are included. The inertia forces
are included with respect to a centre of gravity calculated for the part of the model that is on the
specified side of the sectional plane.

• For panel models the sectional loads from integrating the hydrostatic pressure loads on the panel
model are also reported in the print file (LIS-file).

Note: Sectional loads calculation should be used with care when global matrices are specified
directly as input to Wadam. This is because the global matrices do not provide the
detailed information required to integrate distributed force contributions.

In addition to the total sectional loads it is also possible to compute and write to the Global Results Interface
File (G-file) differnt parts of the sectional load. This includes sectional added mass matrix, sectional damping
matrix, sectional body mass matrix and sectional excitation force. These quantities cannot be handled by
Postresp, but they may be read into DeepC.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 35

To compute these sectional load details a modification of the Wadam*.FEM file must be done manually: Open
this file in an editor an find the card WADAMA1. (This is one of the first cards on the file.) Change the 19th
number on this card from 0 to 1.

2.5.19 Roll Damping Coefficients

Wadam calculates the roll damping coefficient B44 by including contributions from the following hydrody-
namic effects:

• Potential damping from surface wave radiation

• Linearised viscous damping from skin-friction of the hull
• Linearised viscous damping from eddy-making
• Linearised viscous damping from bilge keel
The potential damping effect is included in the radiation potentials and is defined within the regular linear
theory.

The linearized viscous roll damping effects from eddy-making due to the naked hull is computed based on
empirical data given by Tanaka, [6], while the damping coefficient from skin-friction and eddy-making from
bilge keels are computed according to Kato, [7]. These originally non-linear viscous damping contributions
are linearised in order to be included in the harmonic equations of motion.

The potential damping and the skin friction contributions are calculated from a strip model which is auto-
matically generated from a panel model. The other two effects are calculated from information about the
bilge keel.

From version 9.3 of Wadam, a new roll damping model is available which is based on ITTC Recommended
Procedures [14]. The new model is different from Tanaka/Kato’s approach in the method on evaluating the
eddy-making and bilge keel damping components. In addition, a simplified formula is implemented for pre-
dicting the eddy-making component of a sharp-corner box hull with very shallow draught. Moreover, The hull
lift damping component will be added when the forward speed is given for roll damping calculation.

The viscous damping coefficient is generally defined as

∗
B44 = Kη4max

In Wadam the coefficient K is calculated approximately as a function of the wave-frequency, the hull form
and the bilge keel dimensions. The variable η4max is the maximum expected roll angle.
When viscous roll damping is included in the equation of motion the roll angle amplitude is not known in
advance rather it is a part of the solution. Therefore, the user has to estimate the maximum expected roll
angle for each wave heading. The calculated transfer function for roll motion is compared with the maximum
value specified for that heading. If the computed roll motion is ’close’ to the specified angle then the roll
transfer function for that value is assumed to be ’correct’. If the difference is not acceptable then another
η estimate must be made and the roll motion re-computed. Note, however, that the potential solution is
independent of η . Hence the potential should be saved during the first computation and the program should
be submitted with a restart option during this iteration process. If the maximum roll angle is taken from
short term statistics (only one sea state) then Wadam can perform the iteration process automatically. If the
maximum roll angle is taken from long term statistics then the iteration must be carried out manually.

It should be kept in mind that these coefficients are only valid within the range of tests and models used in
the experiments. Extrapolation outside this range should be performed with care.

The roll damping option in Wadam also includes a linearised roll restoring coefficient in the equation of
motion. The calculation of this coefficient is described in Appendix A 3.1.
Note: The Roll damping model can only be used for ship-like structures with symmetry about
the XZ-plane.
Note: When Tanaka/Kato approach is applied, the computation of the bilge keel damping
breaks down when the bilge keel is very small. The symptom is that the damping starts
to increase when the width of the bilge keel decrease. The exact limit is case dependent,
but as an indication the width of the bilge keel should not be smaller than 10 cm.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 36

2.5.20 Global drag-coefficient for roll-damping

A global coefficient for quadratic damping of the roll-motion can be given as user-specified input. This
makes it possible to use results from model-tests directly. The quadratic damping is linearized by stochastic
linearization. This way, model tests performed independent of sea-states can be used in connection with
any given sea-state valid for the actual location of the structure. The formulation of the method of stochastic
linearization is explained in Appendix A 3.5.

Note: This kind of damping should not be combined with any other types of damping effects,
except the potential dampimg. In other words, the old roll-damping model, the Morison
model or user-specified linear damping are not to be combined with the global tortional
drag.

2.6 Calculation Methods

2.6.1 Calculation of Wave Loads from Potential Theory

The potential theory as described in Newman, ref. [1], is applied in Wadam to calculate first order radiation
and diffraction effects on large volume structures. The actual implementation is based on Wamit2 , ref. [2],
which uses a 3D panel method to evaluate velocity potentials and hydrodynamic coefficients.

This implementation can be used for infinite and finite water depths and both single bodies and multiple
interacting bodies can be analysed. The flow is assumed to be ideal and time-harmonic. The free surface
condition is linearised for the first order potential theory while the non-linear free surface condition is im-
posed for the second order potential theory computation. The radiation and diffraction velocity potentials on
the wet part of the body surface are determined from the solution of an integral equation obtained by using
Green’s theorem with the free surface source potentials as the Green’s functions. The source strengths are
evaluated based on the source distribution method using the same source potentials.

The integral equation is discretisized into a set of algebraic equations by approximating the body surface
with a number of plane quadrilateral panels. The source strengths are assumed to be constant over each
panel. Two, one or no planes of symmetry of the body geometry may be present. The solution of the
algebraic equation system provides the strength of the sources on the panels. The equation system, which is
complex and indefinite, may be solved by a direct LU factorization method or by an iterative method.

When using the direct solver, the integral equation is solved directly for each frequency. The time spent
for different wave frequency is virtually equal. In comparison, more iterations are needed for shorter waves
than for long waves when running with iterative solver. In addition, some more effort is required for different
wave headings comparing with the direct solver. Moreover, iterative solver is more sensible to the mesh
3
quality. Generally, the CPU time for direct solver is increasing with panel number following O(N ) whereas
2
iterative solver is following O(N ) , where N is the total panel number in an analysis. It is suggested that
direct solver shall be used when N is less than 10000 and with many wave headings included. Whereas the
iterative solver becomes more favorable when N is more than 15000.

Boundary Value Problem Formulation

The assumption of potential flow allows defining the velocity flow as the gradient of the velocity potential Φ
that satisfies the Laplace equation

∇2 Φ = 0 (2.19)

in the fluid domain. The harmonic time dependence allows defining a complex velocity potential φ related
to Φ by

Φ = Re φ eiωt


(2.20)

where ω is the frequency of the incident wave and t is time. The associated boundary-value problem will
be expressed in terms of the complex velocity potential φ with the understanding that the product of all
complex quantities with the factor eiωt applies. The linearised form of the free-surface condition is

φz − Kφ = 0, on z = 0 (2.21)
2
Wamit is a computer program developed by Massachusetts Institute of Technology.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 37

where K = ω 2 /g and g is the acceleration of gravity. The velocity potential of the incident wave is defined
by

igA cosh (kz + H) −k(x cos β+y sin β)

φ0 = e (2.22)
ω cosh kH
where the wave number k is the real root of the dispersion relation and β is the angle between the direction
of propagation of the incident wave and the positive x-axis.

Linearisation of the problem permits decomposition of the velocity potential φ into the radiation and diffrac-
tion components:

6
X
φ = φR + φD , φR = iω ξj φj , φD = φ0 + φ7 (2.23)
j=1

The constants ξj denote the complex amplitudes of the body oscillatory motion in its six rigid-body degrees
of freedom and φj the corresponding unit-amplitude radiation potentials. The velocity potential φ7 repre-
sents the disturbance of the incident wave by the body fixed at its undisturbed position. The total diffraction
potential φD denotes the sum of φ7 and the incident wave potential.

On the undisturbed position of the body boundary the radiation and diffraction potentials are subject to the
conditions

φjn = nj , φDn = 0 (2.24)

where (n1 , n2 , n3 ) = n and (n4 , n5 , n6 ) = r × n, where r = (x, y, z). The unit vector n is normal to the body
boundary and points out of the fluid domain. The boundary value problem must be supplemented by a
condition of outgoing waves applied to the velocity potentials φj , for j = 1, ..., 7.

The boundary value problems Equation (2.19) - Equation (2.24) are solved by using Green’s theorem to
derive integral equations for the radiation and diffraction velocity potentials on the body boundary. The
integral equation satisfied by the radiation velocity potentials φj on the body boundary takes the form

ZZ ZZ
∂G(ξ; x)
2πφj (x) + φj (ξ) dξ = nj G(ξ; x)dξ (2.25)
Sb ∂nξ Sb

where Sb denotes body mean wet surface. The Green function G(ξ; x) represents the velocity potential at
the field point x due to a point source located at point ξ .

The corresponding equation for the total diffraction velocity potential φD is

ZZ
∂G(ξ; x)
2πφD (x) + φD (ξ) dξ = 4πφ0 (x) (2.26)
Sb ∂nξ

When the fluid velocity components are to be calculated, the integral equation take the following general
form for source instead of velocity potential.
ZZ
∂G(ξ; x) ∂φ(x)
2πσ(x) + σ(ξ) dξ = (2.27)
Sb ∂nx ∂nx

2.6.2 Calculation of Wave Loads from Second Order Potential Theory

The second order theory applied in Wadam is described in [3] and [4]. Wadam calculates sum- and difference-
frequency components of the second order forces, moments and rigid body motions (Quadratic Transfer
Functions) in the presence of bi-chromatic and bi-directional waves.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 38

2.6.3 Removal of Irregular Frequencies

Wadam provides an option to remove the irregular frequencies from the radiation-diffraction solution. This
method is based on extending the integral equation Equation (2.25) and Equation (2.26) to include a panel
model of the interior water plane,
! ZZ ZZ !
2π ∂G(ξ; x) ∂φ(ξ) Sb
φ(x) + φ(ξ) dξ = G(ξ; x)dξ, x∈ (2.28)
−4π Sb +Si ∂nξ Sb ∂nξ Si

where Si denotes the interior free surface. Equation (2.28) indicates a fixed body (or free surface) condition
on the interior free surface panels. The corresponding equation for the source formulation is taking the
following form.
! ZZ !
2π ∂G(ξ; x) ∂φ(x) Sb
σ(x) + σ(ξ) dξ = , x∈ (2.29)
−4π Sb +Si ∂nx ∂nx Si

The panel model of the water plane can be automatically created by Wadam or provided by the user (from
version 9.3). When the automatic meshing option is applied, panels on the interior free surface is appended
to the panel model of each body, and the connectivity between the free surface panels and the waterline
segments is enforced. However, this option should not be used for the body, if a ray from the centroid
of the water-plane area could intersect the waterline more than once, for example, a floater with multiple
columns or a drilling ship with a moon-pool. Instead, a separated panel model for the interior water-plane
area should be provided. Such a panel model can be included as a dummy body in a multi-body analysis.
One may check Section 2.1.6 to determine the order of the bodies when this option is used. It should be
noted that the connectivity between the free surface panels and those on the wet hull surface on waterline
segments is recommended but not mandatory.

2.6.4 Damping free surface lid

It is known that linear potential flow theory over-predicts the free surface elevations in a confined water-
plane area, for example, the gap between two ship-hulls or the moon-pool area of a drilling ship. Wadam
(from version 9.3) provides an option to suppress the unrealistic free surface elevations with an panel model
placed on the confined water-plane area with a imposed damping free surface condition as following.

φn = Kφ(1 − 2i − 2 ), on Se (2.30)

Comparing with Equation (2.21), the additional terms with linear damping factor  are added on the exterior
free surface area Se .

The integral equation Equation (2.25) and Equation (2.26) are extended to include a panel model of the
exterior water plane where the linear damping is imposed.
! ZZ ZZ ZZ !
2π ∂G(ξ; x) ∂G(ξ; x) ∂φ(ξ) Sb
φ(x)+ φ(ξ) dξ+K(2i+2 ) φ(ξ) dξ = G(ξ; x)dξ, x ∈
4π Sb ∂nξ Se ∂nξ Sb ∂nξ Se
(2.31)

The corresponding equation for the source formulation is taking the following form.
! ! !
ZZ ∂φ(x)
2π ∂G(ξ; x) ∂nx Sb
σ(x) + σ(ξ) dξ = , x∈ (2.32)
4π Sb +Se ∂nx Kφ(1 − 2i − 2 ) Se

The panel model for the damping free surface lid can be included as a dummy body in a multi-body analysis.
One may check Section 2.1.6 to determine the order of the bodies when this option is used. The linear
damping factor  is determined by the geometry of the confined water area and can be obtained from model
test.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 39

2.6.5 The Equation of Motion

The equation of motion in Wadam is established for harmonic motion of rigid body systems expressed in the
global coordinate system.

By applying Newtons law and including the added mass, damping and exciting force contributions acting on
the panel and Morison parts of a hydro model the complex 6 by 1 motion vector X(ω, β) can be found from
the equation of motion
 2 
−ω (M + A(ω)) iω (B(ω)p + Bv ) + C + Ce X(ω, β) = F(ω, β) (2.33)

where

M represents the 6 by 6 body inertia matrix

A(ω) represents the 6 by 6 frequency dependent added mass matrix

B(ω)p represents the 6 by 6 frequency dependent potential damping matrix

Bv represents the 6 by 6 linearised viscous damping matrix

C represents the 6 by 6 hydrostatic restoring matrix

Ce represents the 6 by 6 external restoring matrix

F(ω, β) is the 6 by 1 complex exciting force vector for frequency ω and incident wave
heading angle β

The eigenvalues λ and eigenvectors Φ of the rigid body system is obtained for a given incident wave fre-
quency by solving the eigenvalue problem

[−λ (M + A(ω)) + C] Φ = 0 (2.34)

The natural periods of the rigid body system at a given incident wave frequency is expressed as

2π
T =√ (2.35)
λ

2.6.6 Morison’s Equation

Morison’s equation is used in Wadam to calculate contributions to the equation of motion Equation (2.33)
and to calculate the detailed forces F acting on 2D Morison elements and 3D Morison elements. The form
of Morison’s equation used in this calculation is given in Equation (2.36) with the effect of relative motion
included.

F = ω 2 (M + ρVM Ca ) − ω 2 ρVM (Ca + I) x + iωB (x − ξ) + fc + fg + fb (2.36)

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 40

where

ω Incident wave frequency

M 3 by 3 diagonal mass inertia matrix

Ca 3 by 3 diagonal added mass coefficient matrix

I 3 by 3 identity matrix
ρ Density of water

VM Displaced volume of the Morison element

1 8
B Linearised viscous damping matrix expressed as B = 2 ρσCD · 3π Vmax

CD 3 by 3 diagonal drag coefficient matrix

σ Projected area of the Morison element

x Complex amplitude of the incident wave field

ξ Complex amplitude of the motion

fc Fluctuating hydrostatic restoring force representing the first order restoring contribu-
tions integrated in the equation of motion

fg Fluctuating gravity force representing the acceleration of gravity calculated in a coor-

dinate system fixed with the Morison model
fb Fluctuating buoyancy force calculated in a coordinate system fixed with the Morison
model

The linearised viscous damping matrix B in Morison’s equation (Equation (2.36)) is obtained from linearisa-
tion of the general viscous drag force FD expressed as

1 1 8
FD = ρσCD (v − ẋ) |v − ẋ| = ρσCD Vmax (v − ẋ) = B (v − ẋ) (2.37)
2 2 3π

The term
8
Vmax (2.38)
3π

is a standard result obtained by assuming equal work done at resonance by the non-linearised and the
equivalent linear damping term. Vmax is a linearising velocity amplitude specified as input to Wadam. The
one and same Vmax is applied in the linearised drag force calculation for all the motion modes and for all the
incident wave frequencies.

For deterministic analysis of fixed structures Morison’s equation is expressed on the form

1
F = ρVM (1 + Ca ) v̇ + ρσCD v|v| (2.39)
2

where time independent current may be included in v.

The contributions from Morison elements are calculated in the particular element local coordinate systems.
The contributions are transformed into the body coordinate system prior to the assembling of rigid body
quantities.

The wave kinematics in Equation (2.36) and Equation (2.39) may be taken from the incident wave field as
described in Section 2.4.1. Optionally the wave kinematics may also be taken from the diffracted wave field
calculated from potential theory.

2.6.7 Calculation of Tank Pressures, Quasi-static Method

Wadam calculates by an approximate algorithm the harmonic pressure load on the wet finite element sides
in internal tanks.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 41

The loads are calculated by applying a hydrostatic pressure distribution in the accelerated reference frame
fixed with respect to the tank. The pressure load is divided in a constant and an oscillating part and repre-
sented by separate load cases. The pressure gradient is given by

∇p = ρ (g − a) (2.40)

where g is the acceleration due to gravity, a is the complex acceleration of the mid-point of the tank and ρ is
the mass density of water. The gravity vector described in a coordinate system oscillating with the body has
a constant and an oscillating part. Accordingly, the pressure gradient described in the body-fixed coordinate
system has a constant part ρg and an oscillating (dynamic) or fluctuating part:

∇pf = ρ (gf − a) (2.41)

where gf is the fluctuating part of gravity. A detailed outlining of the tank pressure calculation is included in
Appendix A 3.4.

The tanks are modelled by defining wet surfaces for the tanks, in separate load cases. See also Section 2.1.8.
The number of the loadcase will be one higher than the number of the tank. The loadcase numbering must
start at 2, since the wet surfaces with load case 1 are for the external pressure. All surfaces with the same
load case number will be assumed by Wadam to be part of the same tank.

The filling of the tanks is controlled by assigning the hydro-pressure load only to the wet part of the tank
walls. No sloshing effects are included in this formulation and no moments of inertia about the center of
mass of the fluid are accounted for.

By default, the gradient-value is computed in an approximation to the centroid of the tank and the point of
zero pressure is taken as an aproximated centroid of the "roof "of the tank. These reference-points may, in
some cases, not be adequate to use. Because of this, the user has the possibility to specify on input, the
preferred coordinates for gradient-point and point of zero for each tank. The coordinates of the points are
given in the input system of the model. The user may choose, whether only the gradient-point, only the
point of zero, or both are to be given as input.
Note: The mass of the tank fluid must be included in the mass model for Wadam. In the
structural (FE) model the inertia forces from the tank fluid are represented as pressure
loads and should therefore not be included in the structural mass.

2.6.8 Calculation of Tank Pressures, Full Dynamic Method

As an alternative approach to the quasi-static method, the full dynamic method can be selected to solve the
potential flow problem in the isolated fluid domain for each tank (see [10], [11] and [12]) .
The added mass and restoring stiffness of each tank enter into the global motion equation and the radiation
potentials with the resulting motions determine the actual pressure loads in the tanks with the given internal
free surface level:

∇p = −ρ (Φi + gz) (2.42)

in which, Φ is the total radiation potential in a tank.

To calculate the internal tank fluid dynamics, tank definitions shall be included in the panel model, following
the same practice as for the tanks in the structure model. The geometry, wet surface, and the load case
numbers shall correspond to those in the structure model, so that correct load transfer mapping can be
secured.

Tanks are classified into three types: empty, partially filled and fully filled. The empty tanks are bypassed in
the calculation. For a partially filled tank, the zero point of the hydrostatic pressure is, by default in HydroD,
put to the mean internal free surface level. The adjustment of the panels at the internal tank surface is
carried out in the same way as for the external wet panels. For a fully filled tank, the zero point of the
hydrostatic pressure is, by default in HydroD, at the top point of the tank geometry. It may also, by the user,
even be defined above it. A panel element located at the roof center is found by Wadam to have the zero
dynamic pressure for a fully filled tank. It should be noted that, different from the quasi-static method, the
acceleration reference point (gradient point) is not used.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 42

Moreover, the tank fluid shall be excluded from the mass model, since it is represented by the added mass
from the internal radiation problem.

Wadam version 9.0.4 assumes the internal tanks take the same symmetry property as for the external wet
hull. From version 9.1 of Wadam, the symmetry could be different between tanks and external hull. It should
be noted, HydroD Version V4.6-03 prepares the input to Wadam in such a way that the symmetry property
for all tanks follow the symmetry property for external wet hull. However, With version newer than V4.6-03,
HydroD always assumes that there is no symmetry at all for tanks, so that the complete geometry shall be
given for each tank in the panel model.

2.6.9 Pressure Loads up to Free Surface

Wadam may be used to extrapolate to the free surface the panel pressures calculated by first order potential
theory (up to the still water level). Note that this implies that the dry finite elements below the still water
level will receive no loads when the free surface is below the still water level as shown in Figure 2.26.
A constant extrapolation, also called stretching, of pressures above the still water level is applied. This
pressure extrapolation option depends on the amplitudes of the waves and hence this option is only available
when results in the deterministic output format is specified.

Figure 2.26: Constant extrapolation of panel pressures to the free surface

Note that the pressures only will be received by the finite elements above the still water level defined as
wet in the structural model.

For a panel model this option is available for both fixed and floating structures, but the intersection of the
structure and the free surface must be vertical. If a Morison model is included the option is only available
for fixed structures.

2.6.10 Reduced pressure up to the free surface

One of the "stretching" methods is the approach referred to as reduced loads. This is a kind of "mean-
stretching" applied in frequency domain.

In this procedure, the loads are multiplied with a linearly attenuated reduction factor in a zone between
some given distance below and above the still-water level. Below the still-water level, the reduction factor
is applied directly. Above the still-water level, the reduction factor is applied to the loads extrapolated from
just below (ideally at) the still-water level.

The loads are attenuated linearly down to zero as the level increases from still-water level and up to a given
amplitude. This amplitude is typically an extreme-level of the waves (eg a 20 year level). For an amplitude
A the pressure load on a panel with centroid at level zc above the still-water level zw are then:

PR = wP0 ; zw ≤ zc ≤ A (2.43)

where P0 is the pressure at the panel-centroid nearest to still-water level approximately vertically below the
panel with centroid at zc and w is the linear attenuation factor defined by
A − (zc − zw )
w= (2.44)
2A

The reduction is continued down below the still-water level, but is then applied to the pressure at the panel
-centroid:

PR = wP (zc ); A ≤ (zc − zw ) < 0 (2.45)

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 43

where P (zc ) is the pressure at the panel-centroid at zc . P (zc ) and P0 are the standard pressure-calculations
from the boundary value problem bounded by the still-water level.

The pressure reduction procedure is only applicable to vertical, or nearly vertical, sides of the body, typically
the mid-section of a ship hull. The limits in x-direction, of the region we want to apply pressure-reduction on
can be given as input. If no limits are given the program tries to perform pressure reduction on the whole
body.

2.6.11 Wave-current interaction

Based on the perturbation expansion used for the wave-drift damping calculation (see [9]), results for the
RAO’s for a small current or forward speed, are obtained. Added mass and damping matrix and fluid velocity
at off-body points are still presented only for zero speed. This feature can not be combined with dynamic
tank pressure approach in the same analysis. A comparative study between Wadam and two other computer
programs is presented in [13], where it is indicated that Wadam can be applied to the conditions with current
speed up to around 1-2 m/s.

2.7 The Save-Restart System

The save-restart option in Wadam provides a mechanism to store potentials from the solution of the ra-
diation and diffraction problem from one Wadam run to the next. Hence, for a given model the radiation
and diffraction potentials for combinations of incident wave frequencies and heading angles need only be
calculated once.
The save-restart file may be viewed as a database for the calculated potentials. That is, potentials for
different combinations of frequencies and heading angles may be appended to the save-restart file from a
sequence of runs. Furthermore, Wadam may extract potentials for a subset of the frequencies and heading
angles stored on the file.

If the model is changed the save-restart database (Wadam.RSQ) must be deleted. Otherwise Wadam will not
run. Notice, however, that when Wadam is executed from HydroD this application allows for management
of multiple restart databases.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 44

3 USER’S GUIDE TO WADAM

This chapter describes how to use Wadam to analyse typical hydrodynamic problems involving fixed, float-
ing and tethered structures. Simple tutorial examples as well as more real life engineering problems are
presented. Some practical modelling guidance is also provided.
The examples have been selected to demonstrate the main features of Wadam. They do not cover all options
in the program. The text for each example explains the options used in that specific example only.

The panel models shown in this section may be used as guidance during the modelling phase. It is however
important to create a panel model suitable for the type of analysis to be performed. A panel model that is
suitable for global response analysis is not necessarily acceptable for other types of analyses. This remark
is equally important for the structural model.

3.1 Column-stabilized Unit

A column-stabilized unit can refer to a TLP or a Semi-submersible. The hydrodynamic analysis on a column-
stabilized unit normally involves both potential flow theory applied on a panel model and Morison equations
applied on a Morison model. The following calculations are performed in Wadam for global response analysis.
The first two points are automatically calculated while the rest is optional for this analysis.

• Hydrostatic calculation in which both the hydrostatic and inertial properties for the structure are
calculated

• Calculation of hydrodynamic exciting forces, added mass, damping and global motion response
• Eigenvalue calculations for rigid body motions. One set of eigenvalues is calculated for each wave
period.

• Sectional forces in specified sections. These are used for selection of waves to be used in a structural
analysis.
• Calculation of constant drift forces and moments including the effect of motion. These are used in a
succeeding analysis where the maximum offsets are calculated.
• Calculation of fluid kinematics in specified points used for air gap and wave up welling analyses. The
fluid kinematics include the effect of radiation and reflection of waves from the floater.
In case a structure model is included in the analysis, the following detailed loads will be calculated and
transfered to the structure model.

• Hydrostatic loads
– Gravity acceleration force

– Hydrostatic pressure
– Pretension force acting on anchor/TLP elements

– Buoyancy force on Morison beams and nodes

• Hydrodynamic loads
– Rigid body accelerations and fluctuating gravity
– Hydrodynamic pressure caused by radiation, diffraction and incoming waves

– Restoring pressure induced by the body heave, pitch and roll motions

– Reaction forces on anchor/TLP elements caused by rigid body motions of the floater

– Morison force as line loads on beams where Morison sections are defined
– Fluctuating buoyancy force and incoming wave pressure on nodes when pressure area ele-
ments are defined with Morison model
Preparation for this analysis in Wadam consists of the following steps:

• Creation of the panel model in GeniE or Patran-Pre. Symmetry of the structure can be utilized so
that and only one quarter or half of the wet hull is modelled

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 45

 • Creation of beam elements for the Morison model in Patran-Pre or GeniE
• Creation of structural model in Patran-Pre or GeniE
• Including the above Input Interface Files in HydroD, define relevant properties, define mass model
and loading condition
• Specifying environmental properties and execution directives where computation parameters and
options are specified.

Panel Model
The basic part of the panel model for a TLP is shown in Figure 3.1. Since the TLP is double symmetric only
one quarter of the panel model is modelled. The remaining parts of the model are generated in Wadam by
the symmetry option.

Figure 3.1: Basic part of the TLP panel model from two view points

The basic part of the panel model is modelled as one first level superelement in Patran-Pre or GeniE. It is
required that it is modelled in quadrant one, that is with x, y ≥ 0. The input coordinate system is located at
the water level in the centre of the TLP.
The wet sides of the panel model are identified in in Patran-Pre or GeniE by dummy hydro pressure acting
towards the wet hull from the fluid.
The mesh density is important for the hydrodynamic loads. It is of importance that the mesh density re-
flects the hydrodynamic pressure variation around the structure. In areas where the pressure variation is
large element sizes should be small. One may check Section 2.1.2 for more requirements on the mesh
density.

Morison Model
The Morison model is a beam element model; see Figure 3.2 for a TLP. As required by Wadam, the same
input coordinate system as for the panel model is used. Only two node beam elements are accepted as
input for Morison model. Morison beams are included in HydroDin groups identified by beam cross section
property. It is possible to use the same T file as for the structure model, where only part of the beam sec-
tions are taken and rest of the model ignored for Morison model. One may refer to Section 2.1.5 for details.
Only those effective beams are accounted as a part of Morison model. The pressure area elements and
TLP/anchor elements are only allowed to attached on the nodes belongs to the effective Morison beams. All
relevant properties are assigned in HydroD for the Morison cross sections, pressure area elements, TLP/an-
chor elements. One may check Section 2.1.3 carefully when specifying the input parameters for Morison
elements.

The diameter, added mass and drag coefficients of a Morison cross section can be tuned in HydroD for
practical reason. For example, in case the Morison section is defined for columns or braces of the floater
where panel model are also defined, one can decrease the diameter and increase the drag coefficient so
that the product of the them are unchanged. The purpose of this operation is to avoid double counting the
buoyancy force from both the panel model and the Morison model.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 46

In case load transfer is included in the analysis, the line and nodal loads will be transfered to the same beam
or node elements where Morison beams, pressure area elements, TLP/anchor elements are defined.

Figure 3.2: Morison model

Structural Model
The underwater part of the structural model for a TLP is shown in Figure 3.3.

The model is built from first level superelement model created in e.g. Patran-Pre or GeniE. The whole struc-
ture model may contain only one first level superelement model or be built together in the hierarchy with
more than one first level superelement models. In the later case, both the geometry assembly and the de-
scription of load case numbers is built using Presel. The load case number description includes wave loads
from Wadam and loads defined in the pre-processors, see Section 2.1.10 for further description.

Wet surfaces on the structural model are identified in Patran-Pre or GeniE by use of the same modelling
steps as for the panel model. See Section 2.1.8 for further description.

The structural model could also have been used as the hydro model for calculation of hydrodynamic loads,
see Section 2.1.8 for details. Separate models are normally used to optimise panel dimensions and shapes
as well as to minimise the number of panels in the panel model.

Structural mass is distributed with a high degree of accuracy in the model. This is necessary in order to in-
clude the inertia forces, transferred as nodal accelerations, in a proper way in the structural analysis.

Depending on the purpose of the structure analysis, the structural model could be a coarse mesh global
model, or one of the local fine mesh sub-models. The global structural model is used to do load transfer
for a subsequent structural analysis where the ultimate limit state (ULS) capacity of the hull structure is
checked. the local fine mesh sub-model is used for the fatigue limit state (FLS) check. Program Submod can
be used to apply the displacement as the boundary condition for the sub-model from the FE analysis results
of the global model. It is important that the input and results from these analyses are consistent.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 47

 Figure 3.3: The sub-sea part of the structural shell model

Mass Model
When doing a global response analysis, a simple mass model can be adopted in HydroD by specifying total
mass, center of gravity and radius of gyration. When load cross sections are included, it is required that
a distributed mass model shall be used. When a load transfer analysis is performed, the distributed mass
model is often referred to the same T file as used for the global structure model. For detailed description of
these options, one may check Section 2.1.7.

HydroD is very efficient to determine or verify a mass model which is in balance with the buoyancy force (and
pretension force when TLP/anchor elements are specified) from a given loading condition. For any of above
mentioned mass model option, the global mass matrix (and sectional mass matrices when cross sections are
defined with distributed mass model) are computed by HydroD and transferred to Wadam analysis through
WADAM1.FEM file.

Check of Results
During a Wadam analysis, several checks should be performed carefully through the output files, including
the list or print file, G file, L and S files, check Section 4.1.2 for details. The following checks should as a
minimum be made after a global response analysis:
• Error and warning messages
Wadam will give error or warning messages if the input data and the input models contain inconsis-
tencies.

• Hydrostatic properties calculated on the model

All hydrostatic data are printed in the Wadam listing. These properties are more closely explained
in Section 2.5.5.

• Mass properties
A summary of the mass calculations is given in the Wadam listing.
• Rigid body eigenperiods for the structure
The eigenperiods show whether Wadam has interpreted the stiffness and mass properties as ex-
pected and is an extra check of these values.

• Transfer functions for exciting forces, added mass, damping and rigid body motions

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 48

 The transfer functions should generally not be too irregular. Large jumps in the transfer functions,
especially for added mass and potential damping, may be caused by too large wave period steps, the
presence of irregular frequencies or numerical problems in Wadam. The irregular frequencies may
usually be removed by the removal of irregular frequency option, check Section 2.6.3 for details.
In addition to the checkpoints mentioned above, the following points should be checked in the analysis
where the load transfer to a structural model is specified:

• Information about matching between elements in the panel model and the structural model
All wet elements in the structural model should receive wave pressure loads from the analysis. If
the distance from one element in the structural model to the nearest element in the panel model or
the difference in orientation of the element normals is too large then Wadam will not transfer any
loads to the element. A warning will be given in the print file; see Appendix A 3.3. In such cases
the tolerances specified in HydroD should be increased. It is, however, important to be aware of
the possible problems when the tolerances are increased. The transferred pressures may not be
correct for these elements and the load sums for the structural model and the panel model may not
be satisfactory.

• Difference between calculated loads on the panel model and loads transferred to the structural
model

The total pressure loads transferred from the panel model to the structural model are printed in the
Wadam listing. These load sums should be checked against the sum of calculated pressures on the
panel model. A large difference indicates that a new analysis should be performed changing either
the tolerance, the panel model or the structural model. The transfer of pressure loads from the
panel model to the structural model is described in detail in Appendix A 3.3.
• Load sums for the wave loads in the structural analysis
The sum of loads printed in the Wadam listing contains pressure loads on wet elements only. Inertia
forces are not included and thus the total load sums cannot be checked. The total load sums must
therefore be checked in the structural analysis. Since all loads are transferred from Wadam – includ-
ing the restoring loads from the TLP/anchor elements – the load sums for wave load cases should
be close to zero in the structural analysis. Large forces in the load sums indicate errors either in the
load transfer or in the load combinations in Presel.

3.2 Global Response for a Ship

This example illustrates how Wadam may be used in a typical global response analysis for a ship with no
forward speed. This calculation is in principle equal to global response analysis for other types of struc-
tures. The ship used in this example is a typical Aframax tanker with a 62.4 metre long bilge keel on each
side.

This analysis is used for two different purposes:

• Sectional forces are calculated for several sections along the ship. As the ship, in principle, is equal
to a beam with loads sectional forces may be used directly for dimensioning purposes within certain
limits. Combinations of the sectional force components are used to calculate stresses in various
positions of the ship. It should however be remembered that this is a simplification. Structural
analysis should be used for more detailed stress calculations. The sectional forces combination is
performed in the statistical postprocessor Postresp.
• Based on the global responses acceleration components and combined accelerations in specified
points may be calculated in Postresp. This is of less importance for the tanker used in this example
but may be important for other types of ships or barges transporting heavy equipment.

For both alternatives the roll response is of major importance. Non-linear roll damping and restoring are
therefore included according to a linearising procedure. Potential theory is used for all wave periods. For-
ward speed is not allowed in Wadam and is consequently not included.

The two significant differences are the non-linear viscous roll damping and the corrections in restoring forces
due to the non-linear GZ-curve. Both effects are introduced in a linearised form.

The following calculations are performed in Wadam for this analysis. The first two tasks are automatically
calculated while the others are specified for this analysis.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 49

 • Hydrostatic calculation in which both the hydrostatic and inertial properties for the structure are
calculated

• Calculation of hydrodynamic exciting forces, added mass, damping and global motion response for
the ship

• Eigenvalue calculations for rigid body motions.

• Calculation of sectional forces in specified sections
• Calculation of linearised viscous roll damping and restoring terms
Preparations for the analysis in Wadam consist of the following steps:

• Creation of the panel model in Patran-Pre or GeniE. Symmetry of the structure is exploited and only
one half of the ship is modelled.

• Creation of beam elements for the mass model in GeniE or Patran-Pre.

• Including the panel model in HydroD on which the strip model with bilge keel are created for roll
damping evaluation

• Definition of a loading condition in HydroD where the mass model and a GZ-curve is included.
• Definition environmental properties and execution directives where computation parameters and
options are specified.

Panel Model
The total panel model of the ship is shown in Figure 3.4. The basic part of the panel model consists of one
half only. The total model is generated in Wadam through reflection (mirroring) of the basic part.

Figure 3.4: Total panel model of the ship including reflected part

The basic part of the panel model is modelled as one first level superelement in Prefem. The coordinate
system for the model is located at the water level above the centre of gravity of the ship. Modelling advice
for the panel model is given in Section 2.1.2.

Mass Model
The mass model for the ship is shown in Figure 3.5. This is a global response calculation where the lo-
cal mass distribution is modelled in a manner adequate for the sectional force calculations. The mass is
simply modelled as transverse beams representing the mass of each section thereby ensuring that the roll
radius of gyration and metacentric heights are correct. Large additional masses are modelled with point
masses.

Figure 3.5: Mass model for the ship

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 50

Roll Damping Model
For a ship or barge the roll motions will generally be much larger than for floating offshore structures. These
large roll motions increase the viscous damping significantly. The non-linear damping contributions are
therefore included in a linearised manner in Wadam; see Section 2.5.19. The non-linear behaviour of the
roll restoring is also linearised and included in the model. Reference is made to Appendix A 3.1 for further
description. It should be noted that the strip roll damping model described in this section can only be used
for monohull ships. Moreover, a panel model with symmetry about the XZ-plane should be used.
The following non-linear contributions are included in the model:

• Viscous damping from skin friction and eddy-making for the naked hull
HydroD can guide users to create a strip model which uses the panel model as the basis. The strip
model consists of a group of YZ-plane cuts along x axis of the ship. The 2D geometry of the wet
hull in each cut will be used to evaluate the skin friction and eddy-making damping. Integration will
then performed to get the damping over the whole ship length. The length, bilge Radius and Strip
type can be defined in HydroD and transfered to a Wadam analysis. There are following 3 types
option available, see Section 2.5.19 for details. Be noted that the strip type is only relevant when
Tanaka/Kato method is chosen.

2011 ITTC method f o r conventional h u l l

2011 ITTC method f o r barge h u l l
Tanaka / Kato method

• Viscous damping due to the bilge keel

The presence of bilge keel and its location at each strip with its width and angle can be defined in
HydroD interactively. The definition will be transfered to the Wadam analysis.There are two options
as following available for the method to evaluate the bilge keel damping. One may check [14] for
details. Be noted that the options is irrelevant when Tanaka/Kato method is chosen.
Integration along the h u l l
Simplified method

• Viscous damping due to forward speed

It is possible to include the forward speed effect on viscous damping if one of the ITTC methods are
selected, see [14] for details. When a forward speed is given, the lifting damping component will be
calculated. The skin friction and eddy-making damping will also be affected by the forward speed.

• Roll restoring from the GZ-curve

The roll restoring coefficient is modified on basis of the given GZ-curve defined in HydroD. The GZ-
curve to be given as input to Wadam requires equal steps in roll-angle.
When defining the Wadam analysis in HydroD more input parameters for roll damping model are given: flow
type, stochastic linearization options and maximum roll angle. The flow type will determine the Reynold’s
number used for evaluation of the skin friction damping. The non-linear contributions described above
are linearized according to the maximum expected roll angle for each wave heading. When the stochastic
linearization is to be used, the maximum roll angle given in HydroD will be taken as input to the first iteration.
The final roll angles are found from a sequence of analyses performed until convergence between input roll
angles in each iteration and computed extreme roll angles from the sea state specified in HydroD is achieved.
Satisfactory convergence is analysis dependent but should at least be achieved when the error in the angle
is less than 0.1 degrees. Such a stochastic linearization can only be used to find the linearized roll damping
in a short term sea state. In case one want to evaluate the linearized roll damping for a long term sea state,
manual iterations can be performed between Wadam and Postresp analyses.

Check of Results
In addition to the checkpoints mentioned in Section 3.1 for a global response analysis, the following should
be checked in a global response analysis of a ship:

• Roll angles

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 51

 The computed roll RAOs are transferred to Postresp for statistical post-processing. For given sea
state considered, the maximum roll angles should be equal to the converged maximum roll angles
calculated in Wadam.

3.3 Handling of fluid compartments

This example illustrates how fluid dynamics in compartments is handled by using Wadam from HydroD.
Comparisons are made between the quasi-static method and full dynamic method described in Section 2.6.7
and Section 2.6.8. The model used in this example is a 3 meter length floating barge with two rectangle
tanks, see Figure 3.6 for the panel model and layout of the two rectangle tanks. The main particulars of the
barge model can be found in [10]. One should be noted that when fluid dynamics in compartments is to be
accounted for, a structure model, which holds the definitions of compartments, is needed in HydroD even if
a load transfer is not included in the Wadam analysis.

Figure 3.6: Panel and structure model for a floating barge with two compartments

Panel model and structure model

The panel model and structure model are all referred to the same input interface file, where the first load
case LC1 with dummy hydro pressure are applied to the external wet hull surface, dummy hydro pressure
LC2 and LC3 applied to the two tanks respectively. There is no symmetry condition applied, though the
geometry of the model is symmetrical about the XZ-plane. It should be noted that though the symmetric
condition is not allowed to be used for tanks, it can be used for the external wet hull for panel model. For the
quasi-static method, the dummy hydro pressure for tanks are not used for the panel model. In comparison,
radiation potential problems are solved for each tank in addition to the external wet hull when full dynamic
method are used so that the geometry for each tank is needed in the panel model by Wadam. Geometry for
each compartment is identified by dummy hydro pressure with unique load case number. It is practical to
use the same input interface file for the panel model and structure model when the full dynamic approach is
used. They can be different, as long as the geometry and load case numbering for each tank are matching
between the two input interface files.

Mass model
The same mass model for the hull structure is defined in HydroD for both quasi-static method and full dy-
namic method. However, the mass matrix written to WADAM1.FEM (as input to the Wadam analysis) is
different between the two methods. When quasi-static method is used, the mass of the compartment con-
tent is included by HydroD and added to the hull mass, whereas the compartment fluid mass is considered
in terms of tank added mass, and therefore not included in the mass model for the dynamic method.

Restoring matrix due to compartment free surface

The internal free surface of a partially filled tank can bring a negative stiffness for roll and pitch motions.
When quasi-static method is used, such stiffness due to the presence of internal tank free surface is com-
puted, put together for all tanks by HydroD and then transferred to Wadam by a "user specified" restoring
matrix. In comparison, for the full dynamic method, the restoring matrix of each tank is computed in Wadam
instead, as a part of the solution.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 52

Compartment points
The compartment points are acceleration point and zero-level point for each compartment. In most cases,
the acceleration point should be the geometry center of the fluid volume in a compartment, whereas the
zero-level point is the geometry center of the internal fluid free surface in the compartment. These points
can be computed by HydroD (also can optionally be specified or changed by users) provided that the com-
plete geometry of a tank is available in the given structure model. Otherwise, the compartment points shall
be carefully defined by the user. The compartment points are transferred to a Wadam analysis through
WADAM1.FEM file.

When the quasi-static method is being used, both acceleration point and zero-level point are important to
determine the tank pressure to be transfered to the structure analysis. These points are irrelevant in the
case only global response is to be calculated for quasi-static method.

When the dynamic method is being used, only the zero-level point is relevant. Wadam will use this point
to determine the correct compartment free surface level which in turn determines the tank added mass
and restoring matrix. It also determines the compartment pressure distribution for load transfer. Therefore,
zero-level point is a mandatory input for full dynamic method.

Check of Results
The following should be checked when the analysis involves compartment modeling:

• Hydrostatic properties
Make sure that the mass input and restoring coefficients/matrix are as expected.

• Added mass and motions

Response spikes are to be expected when wave frequencies are matching the compartment sloshing
modes. This can be captured by the full dynamic method, but not the quasi-static method. The
RAOs of added mass and motions can be checked in Postresp. One may find [10] for more results
and discussion of this barge model.

• Compartment points
Compartment points are listed in the printing file, check if they are consistent to the input values.

• Compartment pressure distribution in case load transfer is included

The hydrostatic and dynamic pressure can be checked in Xtract. The elements above the zero-level
point should not receive any pressure.

3.4 Free surface lids

A free surface lid is a panel model placed on the free surface either in the interior wet hull domain for
irregular frequencies removal or in the exterior domain to suppress the resonance response of a confined
fluid volume, see Section 2.6.3 and Section 2.6.4 for the theoretical background of the methods.

Dummy hydro model for free surface lid

Each free surface lid is defined as a panel model within a separate dummy hydro model in HydroD. The
mesh should be given in the input coordinate system (the same as for the "regular" bodies). Wadam will
take care of the coordinate translation to the global system. Only a panel model is needed and mass model
is not required. The free surface mesh can be created by HydroMesh from HydroD. By starting HydroMesh
from the loading condition of the hydro model for the ship hull in HydroD, the waterline as the boundary for
the free surface lid is automatically created, as the basis for making the mesh of the lid. The active area can
be shifted in HydroMesh, so that one can select to create mesh on either interior or exterior domain of free
surface for different purpose. The mesh size on the free surface should be comparable to that on the wet
hull.

Body ordering
Free surface lids are included as additional bodies in a multi-body analysis. The following order should be
adopted when defining the multi-body model: floating bodies > fixed bodies > damping surfaces > internal
lids. Damping factor is given with the damping surface when defining the multi-body model. Off-body points
can be defined with the floating body on the damping surfaces to track the free surface elevation.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 53

An analysis is regarded as a hydrodynamic multi-body analysis if more than one regular body is defined.

Side-by-side barges
The example shown in Figure 3.7 demonstrates the usage of free surface lids used for two floating barges of
the same dimension in a side-by-side configuration. The gap between the two barges are covered with
a damping lid and the interior free surfaces are covered with internal lids for irregular frequencies re-
moval.

Figure 3.7: Two rectangular barges in a side-by-side configuration. The meshes for the hulls
are shown in black, internal lid in red and damping lid in blue

Ship with a moon-pool

For a drilling ship with moon-pool, the spikes in RAOs can either come from irregular frequencies or from the
resonance of the moon-pool area. Therefore, to compare analysis results with and without free surface lids
could be instructive. Figure 3.8 shows the panel model for a drilling ship with a circular moon-pool with free
surface lids in both the interior domain and in the moon-pool area.

Figure 3.8: A rectangular ship with a circular moon-pool. The mesh for the hull is shown in
black, internal lid in red and damping lid in blue

Check of Results
The following should be checked when the analysis involves compartment modeling:

• Model properties
Check the list file to make sure that the proper free surface lid types are given. The displaced volume
for a lid body should be zero.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 54

 • First order responses
The responses in motion, added mass and excitation forces can be different between the analysis
results with and without free surface lids. Generally, the spikes from the irregular frequencies can
be removed by including a internal free surface lid for a floating body. Whereas the amplitude of
the response peaks or spikes are reduced when a damping surface is introduced next to the floating
body. When damping surface is defined, the free surface elevations on the confined water area
should be checked.
• Drift forces
As second order responses, drift forces are more sensitive to the irregular frequency and damping
surface than the first order responses. It is possible to get drift forces from the far field integration in
an analysis with free surface lids, as long as there is only one "real" floating body in Wadam analysis.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 55

4 EXECUTION OF WADAM

4.1 Program Environment

Wadam is a batch program. It may be started from a command-window or from HydroD. The input to Wadam
is prepared by HydroD. This program creates the Wadamanalysis control data file WADAMn.FEM as shown in
Figure 4.1.

Figure 4.1: HydroD – Wadam communication

As depicted in Figure 4.1 the entire set of permanent files associated with a Wadam run is:

Analysis control data mandatory input

Input Interface File mandatory input
Print file mandatory output
Hydrodynamic Results Interface File program controlled output
Loads Interface File optional output
Additional analysis control data for Sestra (S-file) program controlled output
Save-restart file for velocity and source potentials optional input/output

In addition to the permanent files Wadam will generate a set of temporary files during the execution. These
files will be opened on the directory where the process is located unless specific assignments are provided
in the command procedures.

HydroD is normally started directly from a shortcut on the desktop.

4.1.1 The Input Files

Analysis Control Data

The file WADAMn.FEM contains all the input data controlling the analysis to be performed by Wadam. The
file is generated and the input data are established by HydroD.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 56

 Figure 4.2: The HydroD GUI and the command for executing Wadam

Input Interface Files

The Input Interface Files contain the model to analyse. The files are established by the pre-processors. A
detailed description of the Input Interface File format may be found in the Input Interface File Description.
The Input Interface Files are formatted files.

There is one Input Interface File for each separate superelement. Wadam imposes the naming convention
that the Input Interface Files must exist in the default directory for the analysis and that there must be no
prefix for the Input Interface Files. That is, the Input Interface File names must be of the form

Tn.FEM

where the T identifies the file as an Input Interface File and n is the superelement number.
With the multi-body option there is no restriction on the file prefix. Hence, with the multi-body option the
Input Interface Files may be of the form
prefixTn.FEM

It should be noted that HydroD (version up to 4.9) will remove the prefix before sending the T files to Wadam.
It is therefore not allowed to use T files with the same super-element number but describing different wet
hull geometries in a multi-body analysis in HydroD.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 57

4.1.2 Output Files

Loads Interface File

The detailed finite element loads from Wadam will be stored in Loads Interface Files; see Section 2.1.8. The
Loads Interface Files are generated for the first level superelements only. The Loads Interface Files constitute
a subset of the Input Interface File data types. The naming convention for Loads Interface Files is similar to
that of the Input Interface Files except that the identifying letter T is replaced by an L. That is:

Ln.FEM

The Loads Interface Files are by default unformatted. They may optionally be specified as formatted.
In addition the utility program Waloco may convert between formatted and unformatted Loads Interface
Files.

If the Loads Interface Files contain dynamic load cases in the frequency domain format an additional analysis
control data file for the subsequent Sestra analysis is created provided that the load case numbering starts
from 1. This is the S-file containing standard analysis control data for the Sestra analysis. The naming
convention for the S-file is

Sn.FEM

where n is the superelement number of the top level superelement in the structural model.

Hydrodynamic Results Interface File

The Hydrodynamic Results Interface File is created if a global response analysis is performed in Wadam. It
contains transfer functions for rigid body responses together with transfer functions for off-body kinematics
and the rigid body matrices. The naming convention for the Hydrodynamic Results Interface File is

G1.SIF for formatted sequential file

G1.SIU for unformatted sequential file
G1.SIN for Norsam direct access format

Print File
The print file contains major results information from a Wadam analysis. The full description of the result
types in the print file is included in Section 2.5 in this manual.

4.1.3 The Save-Restart File

The save-restart file contains the velocity and source potentials obtained from solving the radiation and
diffraction problems. The save-restart file also contains hydro model data used in consistency checking
when the potentials are used in a restart run. The naming convention for the save-restart file is

<name>.RSQ

where the name is specified in HydroDṪhe file is formatted.

4.2 Processor dependent optimization

Utilizing optimized methods for solving sets of linear equations, the computation time is reduced. This
way, the direct solver has improved substantially with respect to computation time. Also, optimization with
respect to parallel processing has lead to higher efficiency.

The user may want to set the environment variable

MKL_NUM_THREADS
and also

NUMBER_OF_PROCESSORS

equal to the number of processors on the computer used. This way, the effect of parallel optimization would
probably best be taken care of.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 58

4.3 Program Requirements

The disk space requirements in Wadam depend on the type of analysis and of the input models. The execu-
tion time is dominated by the solution of the radiation and diffraction problems that is performed for each
incident wave frequency.

4.4 Program Limitations

Wadam imposes restrictions on the size of the hydro model analysis. The size of the structural model used
in the detailed load calculation is virtually unlimited.
The size dependent limitations are listed below. Note however that the limits may vary between different
installations. The actual list of limitations is printed in the status list and in section 1.2 in the Wadam print
file.
The limit values specified in the list below applies to the standard installation:

Geometry limitations:

Maximum number of panels (for the basic part of the model) 50000
Maximum number of free surface panels (for the basic part of the model) 10000
Maximum number of off-body points 2000
Maximum number of nodes in the Morison model 5000
Maximum number of elements in the Morison model 5000
Maximum number of sub-elements in the Morison model when diffracted wave
kinematics is used in Morison’s equation 2000
Maximum number of sub-elements in one Morison element 5
Maximum number of superelements in the Morison model 1
Maximum number of panels corresponding to one Morison sub-element 99
Maximum number of sections for sectional loads 25
Maximum number of bodies 15
Maximum number of tanks 100

Analysis limitations:

Maximum number of wave frequencies 200

Maximum number of headings 73
Maximum number of phase angles with the time output format 72
Maximum number of current profiles 1
Maximum number of levels in one current profile 30

It should be noted that one may get enormous results or error messages from Wadam when several of these
limitations are exploited at the same time.

4.5 Warnings and Error Messages

Warning and error messages may occur in the print file or on the job log file. Error messages will halt the
execution. Warnings are less serious and the execution will continue. However, the warnings may be a
symptom of a more serious problem. Henceforth the program may abort with an error message at a later
stage in the program execution. In any case, warning messages should be carefully interpreted.

Warnings and Error Messages in the Print File

Problems and errors discovered during the check of analysis control data and the input models are reported
on the print file. The input data check includes:

• Check of analysis control data

• Check of input models including the balance between buoyancy and weight of the hydro model
Execution errors due to malfunctioning of a program procedure will result in a short message and a trace
back of all the calling procedures.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page 59

APPENDIX A THEORY

A1 Hydrostatic Forces

A 1.1 Hydrostatic Coefficients

Wadam calculates the non-zero coefficients Cij in the hydrostatic restoring matrix as follows:

C33 = ρgS
C43 = ρgS2 = C34
C53 = −ρgS1 = C35
C44 = ρg (S22 + Vw zB ) − mgzG
C54 = −ρgS12 = C45
C46 = −ρgVw xB + mgxG
C55 = ρg (S11 + Vw zB ) − mgzG
C56 = −ρgVw yB + mgyG

Here:

(xB , yB , zB ) is the centre of buoyancy

(xG , yG , zG ) is the centre of gravity

ρ is the density of the fluid

g is the acceleration of gravity

Vw is the volume of the wetted part of the body

S is the water plane area

R
Si = S0
xi dS i = 1, 2, . . .
R
Sij = S0
xi xj dS i, j = 1, 2, . . .
S0 refers to the body in a static condition.

A2 Morison Element Formulations

The stiffness matrices Km for each anchor or TLP element in a Morison model are accumulated into the
global restoring matrix for the rigid body equation of motion. The Km matrices are established directly
in the motion reference coordinate system and hence no transformations are needed in the accumulation
process. Having solved the equation of motion xg , which represents the global motion of the rigid body
system, the force vector fg for each fairlead node, described in the result reference coordinate system, is
then computed as

fg = −Km xg (A.1)

The stiffness matrix of an anchor or TLP element Km can be written as following.

" #
Rul Rur
Km = (A.2)
Rll Rlr

It is divided into 4 sub 3 by 3 matrices. We may denote the following cross product operators for position
and force.
   
0 −z y 0 −Pz Py
R× =  z 0 −x , F × =  Pz 0 −Px  (A.3)
   

−y x 0 −Py Px 0

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A1

Here:

(x, y, z) is the fairlead position in the global coordinate where the anchor (or TLP) ele-
ment is attached to the floater

(Px , Py , Pz ) is the pretention force at the fairlead point in the global coordinate

Once Rul is determined, the rest 3 sub matrices can be obtained.

Rll = R× Rul , Rur = RTll , Rlr = −R× Rul R× − F× R× (A.4)

Rul is given for an anchor and TLP element as follows respectively.

A 2.1 The Anchor Element Formulation

Following the definition in Equation (A.2), Rul of stiffness matrix Km for an anchor element is defined as
follows:
 
Sh cos2 αx Sh cos αx sin αx 0
Rul = Sh cos αx sin αx Sh sin2 αx 0 (A.5)
 

0 0 Sv

where

Sh is the horizontal spring constant

Sv is the vertical spring constant

αx is defined in Figure A.1

(x, y, z) are the coordinates of the fairlead relative to the motion reference point

Px is the x-component of the pre-tension

Py is the y-component of the pre-tension

Pz is the z-component of the pre-tension

The rest sub matrices can be obtained by Equation (A.4).

Figure A.1: Anchor element definitions

A 2.2 The TLP Mooring Element Formulation

Following the definition in Equation (A.2), Rul of stiffness matrix Km for a TLP element is defined as follows:

 
λ cos2 α + TL sin2 α λ − TL cos α cos β λ − TL cos α cos γ

=  λ − TL cos α cos β λ cos2 β + TL sin2 β λ − TL cos β cos γ 

Rul (A.6)
 

λ − TL cos α cos γ λ − TL cos β cos γ λ cos2 γ + TL sin2 γ

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A2

where

λ is elastic stiffness (force per length)

T is the constant pre-tension

L is the specified tether length at a given offset position

Tx is the x-component of the pre-tension

Ty is the y-component of the pre-tension

Tz is the z-component of the pre-tension

The direction cosines cos α, cos β and cos γ are defined as follows:
x2 − x1 y2 − y1 z2 − z1
cos α = cos β = cos γ = (A.7)
L L L

The separate Km matrices for each TLP node are described with respect to the (x1 , y1 , z1 ) and (x2 , y2 , z2 )
coordinates at the TLP node and the sea-bed, see Figure A.2.

The rest sub matrices can be obtained by Equation (A.4).

Figure A.2: TLP element coordinates

A3 Calculation Methods

A 3.1 Linearisation of Roll Restoring

The linear roll restoring moment is based on the initial metacentric height (GM) calculated from the model
geometry and specified mass properties and is only valid at small heeling angles. From hydrostatics calcu-
lations the GZ-curve is normally known also for large heeling angles. The restoring roll moment at an angle
η4 = φ may be written as

Mr (φ) = ρgV GZ(φ) (A.8)

where V is the submerged volume of the hull and GZ(φ) is the righting arm. The work performed by rotating
the ship through an angle η4 from 0 to φ may be written as:
Rφ Rφ
Wr = 0
Mr (v) dv = ρgV 0
GZ(v) dv = ρgV Ar (φ) (A.9)

where Ar is the area under the righting arm curve as shown in Figure A.3.

At small angles the restoring work may be expanded in a Taylor series giving

Wri = 12 ρgV GZ i φ = ρgV Ari (A.10)

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A3

 Figure A.3: Linearised Roll Restoring Moment from the GZ-curve

where GZ i , the initial moment arm, is related to GM T , the initial transverse metacentric height, as follows:

GZ i = GM T · η̂4 (A.11)

At large angles the linearised roll restoring coefficient is defined as:

C44 = f · GM T (A.12)

where the factor f is defined as the ratio between the area Ar under the GZ-curve from 0 to φ and the area
Ari below a straight line from 0 to φ with slope equal to GZ i /η̂4 = GM T :
Wr Ar 2Ar 2Ar
f= Wri = Ari = GZ i η̂4
= GM T η̂42
(A.13)

A 3.2 Calculation of Line Loads

The line load on a beam finite element is represented as a sequence of constant line load segments. Each
segment will automatically correspond to a sub-element of a 2D Morison element.

The load evaluated at the centre of gravity of each Morison sub-element is represented as constant load
intensities with x, y and z force components acting over the line segments. This is so both for hydrostatic
and hydrodynamic loads. The line loads produced by Wadam do not include any eccentricities which may
exist in the actual load. Figure A.4 shows the line load representation on a beam element with four line
segments.

Figure A.4: Line loads in Wadam

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A4

A 3.3 The Mapping of Loads from Panel Models to Finite Element Models

The hydrodynamic pressure distribution on a panel model is described as a piece wise constant pressure
variation. Each panel is represented with a constant pressure value which is calculated at its centroid. The
mapping of loads from panels to finite elements is based on a minimal distance criteria between the cen-
troids of panels and structural finite elements. That is, each wetted side of a finite element will receive
the constant pressure of the closest panel while satisfying user specified distance and out-of-plane crite-
ria.

The mapping algorithm may be described as: For each element assign normal pressure from the closest
panel (centroid to centroid) provided that Equation (A.14) and Equation (A.15) are satisfied.

Figure A.5 visualises the mapping in a situation where the finite elements are smaller than the panels. The
points Cp and Cfe represent the panel and finite element centroids respectively. The shaded rectangular
elements represent finite elements receiving pressure loads.

Figure A.5: The mapping of loads from panels to finite elements

The functionality of the user specified tolerance parameters DISTOL and ANGTOL, which controls the map-
ping between panels and finite elements may be described as follows.

The panel with its centroid closest to the finite element centroid is a candidate as the source for pressure
transfer to a finite element if the formula
AΣ DIST OL
−1≤ (A.14)
Apanel 100

is satisfied. Here AΣ is the sum of the four shaded triangles Si , for i = 1, 2, 3, 4, shown in Figure A.6 (a)
and Apanel is the area of a candidate panel. If the nearest panel is not accepted by Equation (A.14) then
the program will check, in increasing order of centroid distance, all the 25 closest panels for a panel satis-
fying Equation (A.14). No pressure load is transferred to the finite element for which Equation (A.14) is not
satisfied.

Panels which satisfies Equation (A.14) will be accepted as a source for pressure transfer if the ANGTOL
criteria

φ ≤ AN GT OL (A.15)

is accepted. φ is defined by Figure A.6 (b). Both φ and ANGTOL are given in degrees.

Figure A.6: The DISTOL and ANGTOL functionality

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A5

Wadam reports information from the actual mapping on the form (NDIST, NANG, NOPP) where NDIST shows
the number of the closest panels which has been checked against Equation (A.14) and NANG correspondingly
shows the number of panels which has been checked against Equation (A.15). NOPP shows the number of
panels which has been checked against Equation (A.15) with the modification that φ is replaced by φ +
180. NOPP is indicating that normal vectors of the panel and finite-element are pointing in the opposite
direction.

The number of finite-elements in the structural model with no matching panel is reported by Wadam.

A 3.4 Calculation of Tank Pressures

The calculation of tank pressure by the quasi-static method can be described as follows.The loads are calcu-
lated by applying a hydrostatic pressure distribution in the accelerated reference frame fixed with respect
to the tank. The pressure load is divided in a constant and an oscillating part and represented by separate
load cases. The pressure gradient is given by

∇p = ρ (g − a) (A.16)

where g is the acceleration due to gravity, a is the complex acceleration of the mid-point of the tank and ρ is
the mass density of water. The gravity vector described in a coordinate system oscillating with the body, has
a constant and an oscillating part. Accordingly, the pressure gradient described in the body-fixed coordinate
system has a constant part, ρg, and an oscillating or fluctuating part:

∇pf = ρ (gf − a) (A.17)

where gf is the fluctuating part of gravity.

The mid-point is taken as the centroid of the extreme coordinates of the tank. For a cubic tank the extreme
coordinates are the corners of the tank. If the tank is spherical, the extreme coordinates will be the corners
in a cube circumscribing the sphere. In both these cases the centroid of the corner points will coincide with
the centroid of the tank, but generally, for arbitrary geometrical shapes, it is an approximation.

The oscillating contribution to the total pressure is represented as a complex load case. The zero-level for
the pressure is the mid-point of the top of the tank. This mid-point is approximated by the center of the
rectangle defined by the top surface of the circumscribing cube.
The user may specify other preferred locations of both gradient -point and point of zero for each tank. These
specifications will then take precedence over the points computed by the program.

The pressure loads should preserve load balance. The total force from the fluid on the tank is given by

Z
F= pn dS (A.18)
S

where n is the unit normal vector pointing out of the tank and S is the surface inside the tank. By Gauss
theorem we then have
Z
F= ∇p dV (A.19)
V

where V is the space inside the tank filled with fluid. Substituting for ∇p this gives

F = mf (g − a) (A.20)

mf is the mass of the fluid in the tank. Similarly, the total moment M on the tank can be seen to be given
by

M = xf × F (A.21)

where xf is the position of the centroid of the tank. This shows that the force and moment from the fluid is
balanced by the gravity force on the fluid and the inertia force and moments of the fluid. This is independent
of the reference level of the pressure. Provided an exact load integration and an exact centroid calculation,
the load balance will then be exact if the mass of the fluid is placed in the centroid of the tank in the mass
model.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A6

The calculation of tank pressure by the full dynamic method follows the same procedure as for the quasi-
static method, except that the pressure component with the acceleration shown in Equation (A.16), Equation
(A.17) and Equation (A.20) is substituted with the radiation pressure from the internal potential flow solution.
The acceleration reference point in this case is not necessary to provide.

A 3.5 Global drag-coefficient for roll

The quadratic roll-damping coefficient may be found in model-tests for given ship-hull types, independent
of sea-state. For the actual sea-state in which the ship is to be analysed, stochastic linearization of the
quadratic damping may be performed. The quadratic drag in the roll-motion gives a contribution to the
moment about the x-axis on the form:
(2)
F44 = −B44 |η̇4 |η̇4 (A.22)

(2) (2)
where B44 is the quadratic drag-coefficient and η̇4 is the velocity in roll. Substituting B44 |η̇4 | with a lin-
(1)
earized damping-coefficient B44 , the error random process due to linearization is:

(2) (1)
e(t) = B44 |η̇4 |η̇4 − B44 η̇4 (A.23)

The requirement that the expectation-value of the squared error has a minimum is

(2)
(1) B44 E(|η̇4 |3 )
B44 = (A.24)
E(η̇42 )

where E is the expectation operator.

Obviously:

E(η̇42 ) = ση̇24 (A.25)

Assuming η̇4 is normally distributed, the half-normal distribution for |η̇4 | gives
r
3 8 3
E(|η̇4 | ) = σ (A.26)
π η̇4

where ση̇4 is the standard-deviation of η̇4 . Substituting the two expectation values above in the formula for
(1)
B44 gives the well-known relation for stochastic linearization of a quadratic drag-term:
√ √
(2) 8 3
(1) B44 π ση̇4 8 (2)
B44 = = ση̇4 B44 (A.27)
ση̇24 π

The equation of motion η1 , . . . , η6 , with the stochastically linearized roll-drag included, may be written:

(1)
Mij η̇j + bij η̇j + B44 δij η̇j + cij η̇j = Xi (A.28)

Here, Mij is the sum of the body-mass matrix and added mass matrix. bij is the potential damping and cij
is the restoring. Xi is the excitation-force. An initial estimate of ηj for all wave-frequencies is made, giving
(1)
a corresponding estimate of the standard-deviation and thereby B44 . Then an iteration process is run on ηj
until a reasonable agreement between the estimate of ση̇4 from the previous and next calculation of motion
is obtained. The motion will then be correct in the least square sense.

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page A7

APPENDIX B The Wadam Print File List of Contents

The Wadam print file begins with a list of contents. This list of contents shows which data will be printed
depending on the chosen print switch in HydroD. The list of contents which is common for all analyses is
presented below for reference purposes.
CONTENTS OF THE WADAM LISTING :
−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−

PRINTED RESULTS FOR THE VARIOUS PRINT ALTERNATIVES : ....... 0 1 2 3 4 5

1. CONTENTS AND LIMITATIONS

1.1 CONTENTS OF THE WADAM LISTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
1.2 LIMITATIONS IN THIS VERSION OF WADAM . . . . . . . . . . . . . . . . . . . . . X X X X X X

2. DATA AND MODEL GENERATION

INPUT CARD DECK ( I F BATCH CARD INPUT ) . . . . . . . . . . . . . . . . . . . . X X X X X X
2.1 ANALYSIS CONTROL DATA INCLUDING
− EXECUTION DIRECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− MODEL INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− TOLERANCES AND CRITICAL WAVE LENGTH . . . . . . . . . . . . . . . . . . . . X X X X X X
− CONSISTENCY OF ANALYSIS CONTROL DATA . . . . . . . . . . . . . . . . . . . X X X X X X
2.2 DATA SPECIFYING THE MORISON MODEL INCLUDING . . . . . . . . . . . . . . X X X X X X
− INPUT INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− DUMP OF INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X
− DUMP OF ELEMENT TRANSFORMATION MATRICES . . . . . . . . . . . . . . . . X X
2.3 DATA SPECIFYING THE PANEL MODEL INCLUDING . . . . . . . . . . . . . . . . X X X X X X
− INPUT INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− DUMP OF INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X
2.4 DATA SPECIFYING THE MASS INPUT INFORMATION INCLUDING . . . . . X X X X X X
− INPUT INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− DUMP OF INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X
− DUMP OF MASS DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X
2.5 DATA SPECIFYING THE RESTORING INPUT INFORMATION INCLUDING X X X X X X
− RESTORING DATA FOR EACH INDIVIDUAL INPUT MODEL . . . . . . . . . . . X X X
2.6 CHECK OF DATA ON SAVE FILE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
2.7 SUMMARY OF MODEL PROPERTIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
2.8 ENVIRONMENTAL DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
2.9 MORISON ELEMENT DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
2.10 MORISON NODE DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
2.11 PANEL DATA FOR BASIC PART . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
2.12 PANELS AND MORISON ELEMENTS CORRESPONDANCE DATA . . . . . . . . . . X X X X X
2.13 DEFINITION OF SECTIONS FOR SECTIONAL LOADS . . . . . . . . . . . . . . X X X X X X
2.14 DATACHECK SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X

3. EXECUTION MESSAGES, SAVE/RESTART, VISCOUS FORCES

3.1 COMMENTS ON THE CALCULATION PERFORMANCE . . . . . . . . . . . . . . . . . . X X X X X X
3.2 GENERATION OF SAVE FILE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
3.3 DRAG FORCE LINEARISATION BY ITERATION . . . . . . . . . . . . . . . . . . . . X X X X X X
3.4 VISCOUS ROLL DAMPING MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X

4. PRESENTATION OF RESULTS
4.1 EXPLANATION OF THE RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
4.2 STATIC RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
4.3 GLOBAL HYDRODYNAMIC RESULTS INCLUDING
− FREQUENCY INDEPENDENT MATRICES . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
− FREQUENCY DEPENDENT MATRICES . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
− EIGEN SOLUTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
− RESULTS FROM INDIVIDUAL THEORIES . . . . . . . . . . . . . . . . . . . . . . . X X X X
− TOTAL EXCITING FORCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page B1

 − MOTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
− DRIFT FORCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
− SUM OF DISTRIBUTED LOADS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
4.4 LOAD DISTRIBUTION ON THE MORISON MODEL . . . . . . . . . . . . . . . . . . . X X X X
4.5 PRESSURE DISTRIBUTION ON THE PANEL MODEL . . . . . . . . . . . . . . . . . X X X X
4.6 OFFBODY CALCULATIONS INCLUDING
− FLUID PRESSURE AT SPECIFIED POINTS . . . . . . . . . . . . . . . . . . . . . X X X X X
− FLUID VELOCITIES AT SPECIFIED POINTS . . . . . . . . . . . . . . . . . . . X X X X X
4.7 SECTIONAL LOADS ON THE PANEL MODEL . . . . . . . . . . . . . . . . . . . . . . . X X X X X
4.8 SECTIONAL LOADS ON THE MORISON MODEL . . . . . . . . . . . . . . . . . . . . . X X X X X
4.8b SECTIONAL LOADS ON THE COMPOSITE MODEL . . . . . . . . . . . . . . . . . . . X X X X X
4.9 GLOBAL SECOND−ORDER FORCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X
4.10 SECOND−ORDER PANEL PRESSURE DISTRIBUTION . . . . . . . . . . . . . . . . . X X X X

5. TRANSFER OF RESULTS FOR FURTHER ANALYSIS AND PRESENTATION

5.1 GENERATION OF LOADS INTERFACE FILES INCLUDING
− INPUT INTERFACE FILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− SUBELEMENT LOADS TRANSFERRED TO LOADS INTERFACE FILES . . X X X
− LOADCASE OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
− MATCHING INFORMATION FOR LOADS TO A SHELL / SOLID MODEL. . X X X X
− SUM OF LOADS WRITTEN TO THE LOADS INTERFACE FILE . . . . . . . X X X X X X
− SUM OF CALCULATED LOADS ON THE PANEL MODEL . . . . . . . . . . . . . X X X X X X
5.2 GENERATION OF GLOBAL RESPONSE INTERFACE FILE . . . . . . . . . . . . . X X X X X X
5.3 GENERATION OF GLOBAL RESPONSE FILE (NV1473 ) . . . . . . . . . . . . . . X X X X X X
SUMMARY OF TIMING INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X
SUMMARY OF FILE USAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X X X X X X

|Sesam User Manual|Wadam|[92-7052] www.dnvgl.com/software Page B2

References

[1] Newman J.N. :

"Marine Hydrodynamics"
The MIT Press, 1977 1.1, 2.6.1
[2] WAMIT Users Manual, Version 5.3S,
Department of Ocean Engineering, Massachusetts Institute of Technology 1.1, 2.1.9, 2.6.1

[3] Lee C.-H., Newman J.N., Kim M.-H. and Yue D.K.P. :
"The computation of second-order wave loads"
Published in the OMAE ’91 conference proceedings, Stavanger, Norway 1991 2.1.9, 2.4.1, 2.6.2
[4] The Implementation of Second-order Force Computation in Wadam,
DNV Sesam Report No.: 93-7081, Rev. 0, Oct. 1993 2.6.2

[5] Patel M.H. and Lynch E.J.:

"Coupled dynamics of tension buoyant platforms and mooring tethers"
Eng. Struct., Vol.5, October 1983 2.1.3
[6] Tanaka N.:
"A Study on the Bilge Keel, Part 4. On the Eddy-Making Resistance to the Rolling of a Ship Hull"
Japan Soc. of Naval Arch., Vol. 109, 1960 2.5.19
[7] Kato H.:
"Effect of Bilge Keels on the Rolling of Ships"
Memories of the defence Academy, Japan, Vol. IV, No. 3, pp. 369-384, 1966 2.5.19

[8] Sarpkaya, T. and Isaacson M.:

"Mechanics of Wave Forces on Offshore Structures"
Van Nostrand Reinhold Company, New York, 1981 2.4.1

[9] Finne S., Grue J. and Nestegård A.:

"Prediction of the complete second-order Wave Drift Damping force for offshore structures"
Proceedings, ISOPE 2000, Seattle, 2000 2.5.14, 2.6.11

[10] Ludvigsen A., Pan Z., Gou P. and Vada T.:

"Adapting a linear potential theory solver for the outer hull to account for fluid dynamics in tanks"
OMAE 2013, Nantes, June 2013 2.6.8, 3.3, 3.3

[11] Pan Z., Vada T., Ludvigsen A. and Finne S., DNV, Norway:
"Tank hydrodynamics in the organized analysis by Sesam Manager"
International Conference on Computer Applications in Shipbuilding, Busan, Korea, 2013 2.6.8

[12] Ludvigsen A. and Pan Z.:

"Extensions and improvements to the solution of linear tank dynamics"
OMAE 2015, St. John’s, June 2015 2.6.8

[13] Pan Z., Vada T., Finne S., Nestegård A., Hoff J.R., Hermundstad E.M. and Stansberg C.T.:
"Benchmark study of numerical approaches for wave-current interaction problem of offshore floaters"
OMAE 2016, Busan, June 2016 2.6.11

[14] ITTC:
"ITTC-Recommended Procedures: Numerical Estimation of Roll Damping"
Vol. 7.5-02-07-04.5, pp.1-33, 2011 2.5.19, 3.2, 3.2
ABOUT DNV GL
Driven by our purpose of safeguarding life, property and the environment, DNV GL enables organizations to advance
the safety and sustainability of their business. We provide classification and technical assurance along with software
and independent expert advisory services to the maritime, oil and gas, and energy industries. We also provide
certification services to customers across a wide range of industries. Operating in more than 100 countries, our 16,000
professionals are dedicated to helping our customers make the world safer, smarter and greener.

SOFTWARE
DNV GL is the world-leading provider of software for a safer, smarter and greener future in the energy, process and
maritime industries. Our solutions support a variety of business critical activities including design and engineering, risk
assessment, asset integrity and optimization, QHSE, and ship management. Our worldwide presence facilitates a
strong customer focus and efficient sharing of industry best practice and standards