MIKE 21 Maritime

Frequency Response Calculator and Mooring Analysis

Scientific Documentation

MIKE 2017
DHI headquarters
Agern Allé 5
DK-2970 Hørsholm
Denmark
+45 4516 9200 Telephone
+45 4516 9333 Support
+45 4516 9292 Telefax
[email protected]
www.mikepoweredbydhi.com

m21maritime_scientific_doc.docx/PRB/ALHA/RTJ/SLE/2018-04-16 - © DHI
PLEASE NOTE

COPYRIGHT This document refers to proprietary computer software, which is

protected by copyright. All rights are reserved. Copying or other
reproduction of this manual or the related programmes is
prohibited without prior written consent of DHI. For details, please
refer to your ‘DHI Software Licence Agreement’.

LIMITED LIABILITY The liability of DHI is limited as specified in Section III of your
‘DHI Software Licence Agreement’:
‘IN NO EVENT SHALL DHI OR ITS REPRESENTATIVES
(AGENTS AND SUPPLIERS) BE LIABLE FOR ANY DAMAGES
WHATSOEVER INCLUDING, WITHOUT LIMITATION,
SPECIAL, INDIRECT, INCIDENTAL OR CONSEQUENTIAL
DAMAGES OR DAMAGES FOR LOSS OF BUSINESS PROFITS
OR SAVINGS, BUSINESS INTERRUPTION, LOSS OF
BUSINESS INFORMATION OR OTHER PECUNIARY LOSS
ARISING OUT OF THE USE OF OR THE INABILITY TO USE
THIS DHI SOFTWARE PRODUCT, EVEN IF DHI HAS BEEN
ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. THIS
LIMITATION SHALL APPLY TO CLAIMS OF PERSONAL
INJURY TO THE EXTENT PERMITTED BY LAW. SOME
COUNTRIES OR STATES DO NOT ALLOW THE EXCLUSION
OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL,
SPECIAL, INDIRECT, INCIDENTAL DAMAGES AND,
ACCORDINGLY, SOME PORTIONS OF THESE LIMITATIONS
MAY NOT APPLY TO YOU. BY YOUR OPENING OF THIS
SEALED PACKAGE OR INSTALLING OR USING THE
SOFTWARE, YOU HAVE ACCEPTED THAT THE ABOVE
LIMITATIONS OR THE MAXIMUM LEGALLY APPLICABLE
SUBSET OF THESE LIMITATIONS APPLY TO YOUR
PURCHASE OF THIS SOFTWARE.’

MIKE 2017
CONTENTS
MIKE 21 Maritime
Frequency Response Calculator and Mooring Analysis
Scientific Documentation

1 Introduction ........................................................................................................................1

2 Application Areas...............................................................................................................2

3 Basic Equations .................................................................................................................3

3.1 Geometry and motion of the floating body .......................................................................................... 3
3.1.1 Volume ................................................................................................................................................ 4
3.1.2 Centre of buoyancy ............................................................................................................................. 4
3.2 Equation of motion .............................................................................................................................. 5
3.3 Added mass and damping coefficients ajkt and Kjkt from radiation potential (1st order) ................... 6
3.3.1 Radiation potential............................................................................................................................... 6
3.3.2 Boundary element method .................................................................................................................. 7
3.3.3 Added mass and damping coefficients Aijω and Bijω in the frequency domain ................................. 8
3.3.4 Inverse transforms............................................................................................................................... 8
3.3.5 Response amplitude operators ........................................................................................................... 8
3.4 Wave drift forces (2nd order) ................................................................................................................ 9
3.5 Wave exciting force FjDt ................................................................................................................... 10
3.5.1 Linearization of the Boussinesq wave results ................................................................................... 11
3.5.2 Inverse transforms............................................................................................................................. 12
3.6 Hydrostatic and gravitational restoring coefficients Cjk ..................................................................... 13
3.7 Inertial coefficients Mjk ...................................................................................................................... 13
3.8 External force Fj, nlt ........................................................................................................................... 14
3.8.1 Viscous damping forces Fj, visct........................................................................................................ 14
3.8.2 Mooring forces Fj, moort .................................................................................................................... 15
Mooring lines ..................................................................................................................................... 15
Fenders ............................................................................................................................................. 15
Chains ............................................................................................................................................... 16
3.8.3 Wind forces ....................................................................................................................................... 16
3.8.4 Current forces.................................................................................................................................... 17

4 Numerical Implementation ..............................................................................................18

4.1 Implementation of the theory ............................................................................................................. 18
4.2 Irregular frequencies ......................................................................................................................... 19
4.3 Convergence mode ........................................................................................................................... 20

5 Output Data.......................................................................................................................21
5.1 Frequency Response Calculator ....................................................................................................... 21
5.1.1 Global output files.............................................................................................................................. 21
5.1.2 Vessel-specific output files ................................................................................................................ 22
5.2 MIKE 21 MA ...................................................................................................................................... 22
5.2.1 Global output files.............................................................................................................................. 22
5.2.2 Vessel-specific output files ................................................................................................................ 22

i
6 References ........................................................................................................................24
6.1 Further reading on MIKE 21 MA and its predecessors ..................................................................... 24

APPENDICES

A The Haskind Relation

ii
Introduction

1 Introduction
MIKE 21 Mooring Analysis (MA) is a dynamic mooring analysis software, which uses the
vessel hull shape and associated gyrostatic data to derive the frequency response
characteristics in the frequency domain and uses them to calculate the motions of the
floating vessel under specific (static and/or dynamic) environmental conditions and in
dependence of the mooring arrangement.

Flow, wave, and wind conditions can be provided as two-dimensional data, taking into
account shear currents or complex wave fields, which are found in ports. Thereby, it
takes advantage of the 2D results of the MIKE 21 Hydrodynamic (HD) module and the
MIKE 21 Boussinesq Wave (BW) module.

It solves the equation of motion for the floating body in all six degrees of freedom and
accounts for

• the exact three-dimensional vessel hull geometry and gyrostatic data

• bound and free long-period waves in shallow water
• frictional damping on the fenders in surge, heave, and roll modes
• viscous damping in surge and sway modes
• wind forces
• current forces
• 1st order wave forces
• 2nd order wave drift forces (assuming Newman’s approximation)

This document presents the scientific background for the Mooring Analysis modelling
system, which consists of the two maritime tools; Frequency Response Calculator (FRC)
and MIKE 21 MA. Application areas are outlined in Section 2. The main equations are
given in Section 3, followed by numerical aspects in Section 4. Finally, output data is
specified in Section 5, followed by references and additional literature.

© DHI - Frequency Response Calculator and Mooring Analysis 1

Application Areas

2 Application Areas
The MIKE 21 Maritime models Frequency Response Calculator (FRC) and MIKE 21
Mooring Analysis (MA) are used for the assessment of vessel motions and mooring
forces under specific environmental conditions.

The model can be applied in

• Ports and terminals

• Harbours and marinas
• Offshore environments
- Offshore moorings
- Offshore tandem moorings
- Floating wind turbines
- Ships at anchor

The model can be used directly to evaluate mooring systems with respect to forces and
vessel motions, e.g. to optimise port and offshore terminals, floating storage and
production units, floating wind turbines, etc.

A major application area is the calculation of downtime/operability of a planned terminal.

The situation is depicted in Figure 2.1.

Figure 2.1 Illustration of a typical application area of MIKE 21 MA: vessel serviced at a terminal.

The moored vessel is subject to winds, currents, waves, and drawdown originating from
passing vessels. They can be incorporated in the analyses as time series (e.g. derived
from spectral information, dfs0 files) or as two-dimensional fields (e.g. derived using MIKE
21 HD or MIKE 21 BW, dfs2 files). The vessel motions and the forces in mooring lines,
anchor chains, and fenders are calculated in MIKE 21 MA. The calculated motions are
then to be compared with guidelines on critical motions to assess the terminal operability.
The resulting mooring line forces can be related to maximum breaking loads to assess
the safety for the investigated conditions.

© DHI - Frequency Response Calculator and Mooring Analysis 2

Basic Equations

3 Basic Equations
The solution of the hydrodynamic problem of a floating body subjected to environmental
conditions such as ocean waves or passing vessels can be obtained by applying linear
potential flow theory to the fluid domain. Thereby, surface waves are assumed to be
regular (harmonic) waves of small amplitude. Applying linear boundary conditions on the
fluid free-surface and body-surface is known as the Boundary Value Problem (BVP).
Eventually, the computation of the velocity potential as the solution to the BVP allows
derivation of the hydrodynamic coefficients and excitation forces caused by waves on the
geometry surface (see /4/). The vessel motions are subsequently derived applying linear
equations for floating body’s response in regular waves. This process is illustrated in
Figure 3.1.

Figure 3.1 Flow chart of variables.

In this section, the geometrical conventions are defined in Section 3.1, followed by the
equations of motion in time and frequency domain in Section 3.2, which are populated by
coefficient matrices. Added mass and damping coefficients are derived from the radiation
potentials, which are outlined in Section 3.3, followed by the 2nd order wave drift forces.
The radiation potentials lead also to the formulation of the wave exciting forces, which is
described in Section 3.5. They take further into consideration the linearized incident
Boussinesq wave model data. Sections 3.6 and 3.7 state the hydrostatic and gravitational
resorting coefficients and the inertial coefficients, respectively. In Section 3.8, the external
forces originating from viscous damping, mooring, wind, and current are given.

3.1 Geometry and motion of the floating body

For the floating body motions, the naval architect’s convention is used, which is depicted
in Figure 3.2.

© DHI - Frequency Response Calculator and Mooring Analysis 3

Basic Equations

Figure 3.2 Definition of the six degrees of freedom of a floating body.

The six degrees of freedom are addressed using the unit normal vector 𝑛𝑗 with

• 𝑗 = 1: translation along the 𝑥 axis – surge

• 𝑗 = 2: translation along the 𝑦 axis – sway
• 𝑗 = 3: translation along the 𝑧 axis – heave
• 𝑗 = 4: rotation around the 𝑥 axis – roll
• 𝑗 = 5: rotation around the 𝑦 axis – pitch
• 𝑗 = 6: rotation around the 𝑧 axis – yaw

In the following equations, the origin of the reference frame (0, 0, 0) is located in the
longitudinal centre of gravity, the centreline and at the still water surface. The vertical axis
is pointing positive upwards. The angle of attack of the environmental forcings 𝛽 such as
waves, is defined counter-clockwise with 𝛽 = 0 in positive direction of 𝑥 .

The following sections define the volume of the floating body by the vessel hull surface
and its coordinate of the centre of buoyancy.

3.1.1 Volume

The surfaces of the floating body 𝑆𝑏 obtained from the vessel hull grid file are integrated
in each of the three directions independently

∀= − ∬ 𝑛1 𝑥 𝑑𝑆 = − ∬ 𝑛2 𝑦 𝑑𝑆 = − ∬ 𝑛3 𝑧 𝑑𝑆 (3.1)
𝑆𝑏 𝑆𝑏 𝑆𝑏

where 𝑛1 , 𝑛2 and 𝑛3 are the 𝑥 , 𝑦 and 𝑧 unit vectors normal to the body boundary and the
median of the three volumes.

3.1.2 Centre of buoyancy

The position of the centre of buoyancy is required in the calculations of some of the
components of the hydrostatic restoring matrix. Similarly, the surfaces obtained from the
mesh are integrated accordingly with

© DHI - Frequency Response Calculator and Mooring Analysis 4

Basic Equations

−1
𝑥𝑏 = ∬ 𝑛 𝑥 2 𝑑𝑆 (3.2)
2∀ 𝑆𝑏 1

−1
𝑦𝑏 = ∬ 𝑛 𝑦 2 𝑑𝑆 (3.3)
2∀ 𝑆𝑏 2

−1
𝑧𝑏 = ∬ 𝑛 𝑧 2 𝑑𝑆
2∀ 𝑆𝑏 3 (3.4)

The centre of buoyancy (𝑥𝑏 , 𝑦𝑏 , 𝑧𝑏 ) within the coordinate system shown in Figure 3.2
defines the space for the equation of motion outlined in the following section.

3.2 Equation of motion

The hydrodynamic interaction between the fluid and the floating body is assumed to be
well described by linear potential flow theory. This approach is theoretically valid when

𝑘𝐴
≪1 (3.5)
𝑡𝑎𝑛ℎ(𝑘ℎ)

in which 𝐴 is the wave amplitude, 𝑘 is the wave number and ℎ is the water depth. It is
further assumed that the body motion remains small (ensured by the mooring system),
and estimates for the neglected hydrodynamic effects can be included in the form of
empirical coefficients /2/.

Under these assumptions, the equation of motion can be solved in the time domain and
reads

6
𝑡
∑(𝑀𝑗𝑘 + 𝑎𝑗𝑘 )𝑥̈ 𝑘 (𝑡) + ∫ 𝐾𝑗𝑘 (𝑡 − 𝜏)𝑥̇ 𝑘 (𝜏)𝑑𝜏 + 𝐶𝑗𝑘 𝑥𝑘 (𝑡)
0 (3.6)
𝑘=1
= 𝐹𝑗𝐷 (𝑡) + 𝐹𝑗𝑛𝑙 (𝑡)

The first term at the left-hand side describes the inertia forces, the second term the
hydrostatic forces and the third term the hydrodynamic forces to first order in the body
motion and the wave steepness /2/. They are referred to as impulse-response functions
(IRF’s).

The matrices 𝑀𝑗𝑘 , 𝐶𝑗𝑘 and 𝐾𝑗𝑘 are 6 x 6 matrices of the floating body system. 𝑀𝑗𝑘 and
𝐶𝑗𝑘 are the inertia restoring matrix and the hydrostatic restoring matrices, respectively.
𝑎𝑗𝑘 are impulsive (added mass) contributions, originating from the 𝑡 = 0 limit of the
radiation problem. The forces due to radiated waves generated by the body’s motions are
expressed as a convolution of the radiation impulse response functions, 𝐾𝑗𝑘 /2/.

The right-hand side summarizes all non-linear external forces such as those from the
mooring system and viscous and frictional damping (Froude-Krylov force), 𝐹𝑗𝑛𝑙 (𝑡), and
the wave exciting forces due to scattering of the incident waves, 𝐹𝑗𝐷 (𝑡).

The position and angular rotation of the body in six rigid-body degrees of freedom 𝑥𝑗 (𝑡)
are expressed in Cartesian coordinates, where 𝑥1 = 𝑥 is aligned with the longitudinal

© DHI - Frequency Response Calculator and Mooring Analysis 5

Basic Equations

ship axis pointing forward. The translations are indicated as 𝑥1 = surge, 𝑥2 = sway and
𝑥3 = heave. The rotational motions are 𝑥4 = roll angle, 𝑥5 = pitch angle and 𝑥6 = yaw angle
(refer to Figure 3.2). The over-dot indicates differentiation with respect to time 𝑡 /2/.

The second term in equation (3.6) makes deriving of the matrices 𝐾𝑗𝑘 (𝑡), 𝑎𝑗𝑘 (𝑡)
inefficient in the time domain due to the convolution with time 𝑡. Therefore, the
hydrodynamic calculations are performed in the frequency domain.

For deriving the hydrodynamic coefficients in the frequency domain, the equivalent of the
time-domain impulse response functions (IRF’s) is used. In the context of linear potential
flow theory, this equivalent is obtained by taking the incident wave to be time-harmonic
with the wave frequency, 𝜔, and complex amplitude, 𝜁̃. This leads to the frequency
response functions

∑{−𝜔2 [𝑀𝑗𝑘 + 𝐴𝑗𝑘 (𝜔)] + 𝑖𝜔𝐵𝑗𝑘 (𝜔) + 𝐶𝑗𝑘 }𝑥̃𝑘 (𝜔) = 𝐹̃ 𝑗𝐷 (𝜔), 𝑗 = 1,2, … ,6 (3.7)
𝑘=1

for the hydrodynamic coefficient matrices added mass 𝐴𝑗𝑘 (𝜔), damping 𝐵𝑗𝑘 (𝜔),
hydrostatic and gravitational restoring coefficients 𝐶𝑗𝑘 , and exciting forces 𝐹̃𝑗𝐷 (𝜔).

3.3 Added mass and damping coefficients 𝑎𝑗𝑘 (𝑡) and 𝐾𝑗𝑘 (𝑡) from radiation
potential (1st order)

3.3.1 Radiation potential

Even though part of the surface elevation and flow field is solved by Boussinesq waves in
MIKE 21 MA, integral equations for the flow potential need to be solved to obtain the
potential for radiated waves. This is done in the frequency domain.

The total wave potential can be decomposed into incident (𝑗 = 0), scattered (𝑗 = 7) and
radiated wave modes (𝑗 = 1,2, … ,6) respectively

𝜑 = 𝜑0 + 𝜑7 + 𝑖𝜔 ∑ 𝜉𝑗 𝜑𝑗 (3.8)
𝑗=1

𝜉𝑗 refers to the six degree of freedom response of the floating body, thus 𝜑𝑗 has a
different unit compared to the other two components. It can be understood as the wave
potential per unit velocity of the floating body.

The velocity potential for the incident waves is defined as

𝑖𝑔𝐴 cosh[𝑘(𝑧 + ℎ)] −𝑖(𝑘𝑥𝑐𝑜𝑠𝛽−𝑘𝑦𝑠𝑖𝑛𝛽)

𝜑0 = 𝑒 (3.9)
𝜔 cosh(𝑘ℎ)

with the coordinates (x, y, z). Note that 𝑧 is related to the water surface.

On the surface of the floating body, the condition in the expression

𝜕𝜑7 𝜕𝜑0
=− (3.10)
𝜕𝑛 𝜕𝑛

© DHI - Frequency Response Calculator and Mooring Analysis 6

Basic Equations

has to be satisfied. For all eight modes 𝑗 = {0,1, … ,7}, the linearized form of the surface
boundary condition is implemented at the surface, where 𝑧 = 0 as shown in the equation

𝜔2 𝜕𝜑𝑗
− 𝜑𝑗 + =0 (3.11)
𝑔 𝜕𝑧

Using the boundary conditions, the Laplacian of the wave potential has to be solved for
all 8 modes (where 𝑗 = {0,1,2, … ,7}); one each for incident and scattered, and six for
each degree of freedom of the radiated waves

∇2 𝜑𝑗 = 0 (3.12)

3.3.2 Boundary element method

The frequency boundary value problems governed by the Laplace equation (3.12) are
solved to obtain the radiation potential φj on the surface of the floating body

𝜕𝜑𝑗 (𝑥𝑖 )
= 𝑛(𝑥𝑖 ) (3.13)
𝜕𝑛

with the normal velocity for each mode 𝑛. When written on the control point of each panel
of the body, the integral equations of the boundary element method are transformed into
a linear system of equations

𝑁
𝜎𝑅 (𝑥𝑖 ) 𝜕𝜑𝑗 (𝑥𝑖 )
+ ∑ 𝜎𝑅 (𝑥𝑘 )𝐾𝑖𝑘 = (3.14)
2 𝜕𝑛
𝑘=1

The integral equations use a source distribution in order to solve the source strengths σR
by calculating the radiation impulse response functions in the 𝑁 × 𝑁 matrix

1 𝜕
𝐾𝑖𝑘 = − ∬ 𝐺(𝑥𝑖 ; 𝑥𝑘 )𝑑𝑆(𝑥𝑘 ) (3.15)
4𝜋 𝜕𝑁𝑥𝑖 𝑆

In this equation, G(𝑥𝑖 ; 𝑥𝑘 ) is the free-surface Green’s function with 𝑥𝑖 and 𝑥𝑘 , which are
the field and source points, respectively /1/.

The radiation potential can then be derived using

𝜑𝑗 (𝑥𝑖 ) = − ∑ 𝜎𝑅 (𝑥𝑘 )𝑆𝑖𝑘 (3.16)

𝑘=1

where

1
𝑆𝑖𝑘 = ∬ 𝐺(𝑥𝑖 ; 𝑥𝑘 ) 𝑑𝑆(𝑥𝑘 ) (3.17)
4𝜋 𝑆

© DHI - Frequency Response Calculator and Mooring Analysis 7

Basic Equations

3.3.3 Added mass and damping coefficients 𝐴𝑖𝑗 (𝜔) and 𝐵𝑖𝑗 (𝜔) in the frequency
domain

The added mass and damping coefficient matrices are obtained from the radiation
potentials 𝜑𝑗𝑅 as defined by

𝑖
𝐴𝑖𝑗 − 𝐵 = 𝜌𝑔 ∬ 𝑛𝑖 𝜑𝑗𝑅 𝑑𝑆 (3.18)
𝜔 𝑖𝑗 𝑆𝑏

The real and imaginary part of this equation can be expressed by

𝐴𝑖𝑗 (𝜔) = 𝑅𝑒 {𝜌 ∬ 𝜑𝑗𝑅 𝑛𝑖 𝑑𝑆} (3.19)

𝑆𝑏

and

𝐵𝑖𝑗 (𝜔) = 𝐼𝑚 {𝜌 ∬ 𝜑𝑗𝑅 𝑛𝑖 𝑑𝑆} (3.20)

𝑆𝑏

3.3.4 Inverse transforms

Once 𝐴𝑗𝑘 is obtained, the added mass coefficients 𝑎𝑗𝑘 can be calculated from

1 ∞
𝐴𝑗𝑘 (𝜔) = 𝑎𝑗𝑘 − ∫ 𝐾 (𝑡) 𝑠𝑖𝑛𝜔𝑡 𝑑𝑡 (3.21)
𝜔 0 𝑗𝑘

As 𝜔 → ∞, the second term vanishes. To obtain the damping coefficients, an inverse

transform of 𝐾𝑗𝑘 (𝑡) is required as shown in

∞
𝐵𝑗𝑘 (𝜔) = ∫ 𝐾𝑗𝑘 (𝑡) 𝑐𝑜𝑠𝜔𝑡 𝑑𝑡 (3.22)
0
(see /5/).

3.3.5 Response amplitude operators

The response amplitude operators (RAO) are not directly provided. They are obtained by
equation

∑ 𝜉𝑗 [−𝜔2 (𝑀𝑖𝑗 + 𝑎𝑖𝑗 ) + 𝑖𝜔𝑏𝑖𝑗 + 𝑐𝑖𝑗 ] = 𝐴𝑋𝑖 (3.23)

𝑗=1

where 𝜔 is the angular wave frequency, 𝜉𝑗 is the body motion, 𝑎𝑖𝑗 , 𝑏𝑖𝑗 and 𝑐𝑖𝑗 are added
mass, damping and restoring matrices, respectively. 𝑋𝑖 is the wave excitation force per
unit wave amplitude, and 𝐴 is the wave amplitude.

The equation is solved for the body motion 𝜉𝑗 , and the complex RAO for the 𝑗th mode is
obtained by

© DHI - Frequency Response Calculator and Mooring Analysis 8

Basic Equations

6
𝜉𝑗 −1
𝑍𝑗 (𝜔, 𝜃) = = ∑[𝐶𝑖𝑗 ] 𝑋𝑖 (3.24)
𝐴
𝑗=1

−1
where [𝐶𝑖𝑗 ] indicates the inverse matrix of the member between the brackets in the
equation of motion. The RAO are given for each angular wave frequency 𝜔 and incident
wave direction 𝜃 /10/.

𝐴 = 𝐴(1) 𝑐𝑜𝑠(𝜔𝑡) + 𝐴(2) 𝑠𝑖𝑛(𝜔𝑡) = ℜ(𝐴̃𝑒 −𝑖𝜔𝑡 ) (3.25)

where 𝐴̃ = 𝐴(1) + 𝑖𝐴(2) . The imaginary part of the harmonic motion is defined as phase
lag relative to the real component /10/.

3.4 Wave drift forces (2nd order)

Second order wave drift forces apply the Newman’s approximation /8/ and are derived by
solving the source distribution for the diffraction problem, 𝜎𝐷 , in each incident wave
direction 𝛽

𝑁
𝜎𝐷 (𝑥𝑖 ) 𝜕𝜑0 (𝑥𝑖 )
+ ∑ 𝜎𝐷 (𝑥𝑘 )𝐾𝑖𝑘 = − (3.26)
2 𝜕𝑛
𝑘=1

In the frequency domain, the second order drift forces are computed by the far field
formula /13/. The mean surge and sway forces

2
𝑥 𝑘 ∗ (𝑘𝑖𝑛𝑓 ∗ ℎ)
𝐹 ( ) (𝛽) = −2𝜋𝜌 2
𝑦 ℎ ((𝑘 ∗ ℎ)2 − (𝑘 ∗ ℎ) + 𝑘 ∗ ℎ)
𝑖𝑛𝑓 𝑖𝑛𝑓
2𝜋 (3.27)
𝑐𝑜𝑠(𝜃) cos(𝛽)
∗ ∫ |𝐻(𝜃)|2 ∗ ( ) 𝑑𝜃 − 2𝜋𝜌𝜔 ∗ ( )
0 𝑠𝑖𝑛(𝜃) sin(𝛽)
∗ 𝐼𝑚(𝐻(𝛽))

with the wave propagation direction 𝛽 , the water density 𝜌, the wave number 𝑘 , the water
depth ℎ, the incident wave direction 𝜃, the infinite depth wave number 𝑘𝑖𝑛𝑓 , and the
Kochin function 𝐻 from 0 to 2𝜋 /5/.

The mean yaw moment

2
𝜌𝜔 2𝜋𝜌(𝑘𝑖𝑛𝑓 ∗ ℎ)
𝑀𝑧 (𝛽) = 2𝜋 ∗ ∗ 𝑅𝑒(𝐻′ (𝛽)) − 2
𝑘 ℎ ((𝑘 ∗ ℎ)2 − (𝑘 ∗ ℎ) + 𝑘 ∗ ℎ)
𝑖𝑛𝑓 𝑖𝑛𝑓 (3.28)
2𝜋
∗ 𝐼𝑚 (∫ ̅̅̅̅̅̅̅
𝐻(𝜃) ∗ 𝐻′(𝜃)𝑑𝜃 )
0

where 𝐻 is the Kochin function. It is defined as

6
𝜋 𝜋
𝑖∗
𝐻(𝜃) = 𝐻𝐷 (𝜃) ∗ 𝑒 2 + 𝑖𝜔 ∗ ∑ 𝑍𝑗 ∗ 𝐻𝑅 𝑗 (𝜃)𝑒 −𝑖 2 (3.29)
𝑖=1

© DHI - Frequency Response Calculator and Mooring Analysis 9

Basic Equations

where 𝐻𝐷 and 𝐻𝑅 are the diffraction and radiation contributions to the Kochin function,
and 𝑍𝑗 is the response amplitude operator (see section 3.3.5 ). For the radiated wave
modes (𝑗 = 1,2, … ,6),

cosh(𝑘(𝑧 + ℎ)) 𝑘 𝑖(𝑘𝑟−𝜋)

𝜑𝑗 (𝑟, 𝜃, 𝑧) = ∗√ 𝑒 4 𝐻𝑅 (𝜃) (3.30)
cosh(𝑘ℎ) 2𝜋𝑟

For the scattered (diffraction) mode,

cosh(𝑘(𝑧 + ℎ)) 𝑘 𝑖(𝑘𝑟−𝜋)

𝜑7 (𝑟, 𝜃, 𝑧) = ∗√ 𝑒 4 𝐻𝐷 (𝜃) (3.31)
cosh(𝑘ℎ) 2𝜋𝑟

In the time domain, the drift force time series are computed by the method described in
/14/.

It is noted that the assumption of the Newman’s approximation due to the far-field
solution effectively means that only the diagonal terms in the quadratic transfer function
(QTF) is considered corresponding to the monochromatic frequency pairs. This only gives
the mean drift force, and second order effects that are not described by the mean drift
force are not included. This approximation is theoretically valid in deep water conditions,
while in shallow water the full QTF matrix with the off-diagonal terms (bi-chromatic
frequency pairs) should be considered. This is currently not implemented in MIKE 21 MA.

3.5 Wave exciting force 𝐹𝑗𝐷 (𝑡)

The wave exciting force is expressed in terms of incident wave quantities and solutions to
the radiation problem by means of the Haskind relations /11/

∞
⃗ + 𝜌 ∫ ∬ 𝜙̇ 𝐼𝑛 (⃗𝑥, 𝜏) 𝜙𝑗 (⃗𝑥, 𝑡 − 𝜏)𝑑𝑥⃗ 𝑑𝜏
𝐹𝑗𝐷 (𝑡) = ∬ 𝑝𝐼 (⃗𝑥, 𝑡) 𝑛𝑗 (⃗𝑥)𝑑𝑥 (3.32)
𝑆𝑏 −∞ 𝑆𝑏

(see also Appendix A). In this equation, 𝜙𝐼 is a known incident wave potential, which
satisfies the linear free-surface body condition and induces a first order dynamic pressure

𝜕𝜙𝐼
𝑝𝐼 = (3.33)
𝜕𝑡

in the fluid. The incident wave refers to the wave undisturbed by the presence of the
body. A subscript 𝑛 is used to indicate the operation 𝑛 ⃗ , with 𝑛⃗ being the normal vector
⃗ ∙∇
to the equilibrium wetted body surface 𝑆𝑏 , while 𝑛𝑗 is the generalized unit normal in six
degrees of freedom. 𝜙𝑗 is the solution to the 𝑗th mode impulse velocity radiation problem,
which is defined by the boundary condition

𝑛⃗ ∙ ⃗∇𝜙𝑗 = 𝑛𝑗 𝛿(𝑡), 𝑥 ∈ 𝑆𝑏 (3.34)

with the Dirac function 𝛿(𝑡). The first term in equation (3.32) refers to the Froude-Krylov
force, while the second term describes the scattering of the incident wave by the body
when fixed to its equilibrium position /2/.

© DHI - Frequency Response Calculator and Mooring Analysis 10

Basic Equations

The scattering of the incident waves depends on the radiation potential of the floating
body. This potential is derived in the frequency domain for linear incident waves (see
Section 3.3). It is further related to the Boussinesq wave field, which is linearized and
transferred to the frequency domain. This is explained in the following sections.

3.5.1 Linearization of the Boussinesq wave results

The results of a Boussinesq wave simulation have to be linearized to provide incident

wave quantities required by the Haskind relations (see Section 3.5), to obtain the terms
𝑝̃𝐼 (𝑥 , 𝜔) and 𝜙̃𝐼𝑛 (𝑥 , 𝜔) to be used in equation (3.50).

Using the results from the free surface elevation 𝜂(𝑥 , 𝜔) and depth-averaged velocities
𝑢′(𝑥 , 𝑡)and 𝑣′(𝑥 , 𝑡) from the Boussinesq model on a set of points on the intersection of
the still water plane with the floating body 𝑥, the depth-averaged velocity with depth
variation 𝑧 can be expressed as
𝜂
1
𝑢′ = ∫ 𝑢(𝑥 , 𝑧, 𝑡) 𝑑𝑧 (3.35)
𝐷 + 𝜂 −𝐷

and
𝜂
1
𝑣′ = ∫ 𝑣(𝑥 , 𝑧, 𝑡) 𝑑𝑧 (3.36)
𝐷 + 𝜂 −𝐷

In these equations, 𝐷 is the draft of the vessel, 𝑧 is the vertical coordinate pointing
upwards, with 𝑧 = 0 at the water surface, and 𝜂 is the water surface elevation /2/.

Based on the free surface elevation 𝜂(𝑥 , 𝑡) from the Boussinesq model, the dynamic
pressure

𝑝0 (𝑥 , 𝑡) = 𝜌𝑔𝜂(𝑥 , 𝑡) (3.37)

and the vertical velocity component

𝑤0 (𝑥 , 𝑡) = 𝜂̇ (𝑥 , 𝑡) (3.38)

are obtained using the linear free-surface boundary conditions. The subscript 0 indicates
the variables at the free surface /2/.

These quantities in time domain 𝑝0 , 𝑢′ , 𝑣 ′ and 𝑤0 have to undergo Fourier transformation

to obtain the corresponding quantities as a function of the wave frequency
∞
𝐹̃𝑗𝐷 (𝜔) = ∫ 𝐹𝑗𝐷 (𝑡)𝑒 −𝑖𝜔𝑡 𝑑𝑡 (3.39)
0

This transformation results in a series of sine and cosine functions

𝑓𝑛 = 𝑎𝑛 𝑐𝑜𝑠(𝑛𝑥) + 𝑏𝑛 𝑠𝑖𝑛(𝑛𝑥) (3.40)

From linear wave theory, the velocity can be derived using

𝐻 𝑔𝑇 𝑐𝑜𝑠ℎ[𝑘(𝑧 + ℎ)]
𝑢= 𝑐𝑜𝑠𝜃 (3.41)
2 𝐿 𝑐𝑜𝑠ℎ(𝑘ℎ)

© DHI - Frequency Response Calculator and Mooring Analysis 11

Basic Equations

The hyperbolic cosine terms are used in the equations

cosh[𝑘(𝑧 + ℎ)]
𝑝(𝑥 , 𝑧, 𝑡) = 𝐹 −1 {𝑝̃0 (𝑥 , 𝜔) } (3.42)
cosh(𝑘ℎ)

cosh[𝑘(𝑧 + ℎ)]
𝑢′(𝑥 , 𝑡) = 𝐹 −1 {𝑢̃′ (𝑥 , 𝜔) } (3.43)
cosh(𝑘ℎ)

cosh[𝑘(𝑧 + ℎ)]
𝑣′(𝑥 , 𝑡) = 𝐹 −1 {𝑣̃ ′ (𝑥 , 𝜔) } (3.44)
cosh(𝑘ℎ)

and

sinh[𝑘(𝑧 + ℎ)]
𝑤(𝑥 , 𝑧, 𝑡) = 𝐹 −1 {𝑤
̃ 0 (𝑥 , 𝜔) } (3.45)
sinh(𝑘ℎ)

to include the hyperbolic depth variations. The linear dispersion relation

𝜔2 = 𝑔𝑘 tanh(𝑘ℎ) (3.46)

is utilized to determine the wave number 𝑘 . It should further be noted that equations
(3.43) and (3.44) apply the depth-averaged velocities directly, translating them directly to
the frequency domain, denoted as 𝑢̃′ and 𝑣̃ ′ . They are used to derive the velocity at the
intersection point between the structure and the still water surface. The relation is given in
the equations (3.35) and (3.36). They can be rewritten as

𝑘ℎ
𝑢̃0 (𝑥 , 𝜔) = {𝑢̃′ (𝑥 , 𝜔) } (3.47)
tanh(𝑘ℎ)

𝑘ℎ
𝑣̃0 (𝑥 , 𝜔) = {𝑣̃ ′ (𝑥 , 𝜔) } (3.48)
tanh(𝑘ℎ)
𝑘ℎ
The term approaches unity in shallow water due to no variations in depth, and 𝑘ℎ
tanh(𝑘ℎ)
in deep waters /2/.

The quantity 𝑝̃0 is the incident pressure acting on the wetted body surface 𝑝̃𝐼 . The
normal derivative of the incident velocity potential 𝜙̃𝐼 (𝑥 , 𝜔) can be calculated from the
velocities obtained in the three directions with equations (3.45), (3.47) and (3.48) /2/.

The sequence of quantities in this section is shown in the following expression.

𝜂, 𝑢′ , 𝑣 ′ → 𝐹𝑟𝑒𝑒 𝑝0 , 𝑢′ , 𝑣 ′ , 𝑤0 →𝐹𝑜𝑢𝑟𝑖𝑒𝑟 𝑝̃0 , 𝑢̃′ , 𝑣̃ ′ , 𝑤

̃ 0 → 𝐷𝑒𝑝𝑡ℎ 𝑝̃0 , 𝑢̃0 , 𝑣̃0 , 𝑤
̃0
𝑠𝑢𝑟𝑓𝑎𝑐𝑒 𝑡𝑟𝑎𝑛𝑠𝑓𝑜𝑟𝑚
(3.49)
𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛
𝐵.𝐶.

3.5.2 Inverse transforms

Similar to the derivation of 𝐾𝑗𝑘 (𝑡), the exciting force 𝐹𝑗𝐷 (𝑡) is obtained from the inverse
transformation of 𝐹̃𝑗𝐷 (𝜔). It can then be expressed in terms of radiation potentials via the
Haskind relation

© DHI - Frequency Response Calculator and Mooring Analysis 12

Basic Equations

∞
𝐹̃𝑗𝐷 (𝜔) = ∫ 𝐹𝑗𝐷 (𝑡)𝑒 −𝑖𝜔𝑡 𝑑𝑡
0
(3.50)
= ∬ 𝑝̃𝐼 (𝑥 , 𝜔) 𝑛𝑗 (𝑥 )𝑑𝑥 + 𝑖𝜔𝜌 ∬ 𝜙̃𝑗 (𝑥 , 𝜔) 𝜙̃𝐼𝑛 (𝑥 , 𝜔)𝑑𝑥
𝑆𝑏 𝑆𝑏

(see also Appendix A). The input terms 𝑝̃𝐼 (𝑥 , 𝜔) and 𝜙̃𝐼𝑛 (𝑥 , 𝜔) for the Haskind relation
are derived from the Boussinesq wave field, e.g. calculated using MIKE 21 Boussinesq
wave model results (MIKE 21 BW) as described in Section 3.5.1.

3.6 Hydrostatic and gravitational restoring coefficients 𝐶𝑗𝑘

The hydrostatic and gravitational restoring matrix is independent of the wave frequency 𝜔
and only depends on the geometry of the floating body. The matrix in the time domain
𝐶𝑗𝑘 (𝑡) is expressed for the six degrees of freedom (see Figure 3.2) using the relations

𝐶3,3 (𝑡) = 𝜌𝑔 ∬ 𝑛3 𝑑𝑆
𝑆𝑏

𝐶3,4 (𝑡) = 𝐶4,3 (𝑡) = 𝜌𝑔 ∬ 𝑦 𝑛3 𝑑𝑆

𝑆𝑏

𝐶3,5 (𝑡) = 𝐶5,3 (𝑡) = −𝜌𝑔 ∬ 𝑥 𝑛3 𝑑𝑆

𝑆𝑏

𝐶4,4 (𝑡) = 𝜌𝑔 ∬ 𝑦 2 𝑛3 𝑑𝑆 + 𝜌𝑔∀𝑧𝑏 − 𝑚𝑔𝑧𝑔 (3.51)

𝑆𝑏

𝐶4,5 (𝑡) = 𝐶5,4 (𝑡) = −𝜌𝑔 ∬ 𝑥 𝑦 𝑛3 𝑑𝑆

𝑆𝑏

𝐶4,6 (𝑡) = −𝜌𝑔∀𝑥𝑏 + 𝑚𝑔𝑥𝑔

𝐶5,5 (𝑡) = 𝜌𝑔 ∬ 𝑥 2 𝑛3 𝑑𝑆 + 𝜌𝑔∀𝑧𝑏 − 𝑚𝑔𝑧𝑔

𝑆𝑏

𝐶5,6 (𝑡) = −𝜌𝑔∀𝑦𝑏 + 𝑚𝑔𝑦𝑔

The remaining coefficients of the unstated matrix are set to 0. Indices 𝑏 and 𝑔 denote the
centre of buoyancy and the centre of gravity, respectively. 𝜌 is the density of water, 𝑔 is
the gravitational acceleration, 𝑚 is the mass and ∀ is the volume, 𝑆 indicates the surface
of the floating body. 𝑛𝑗 is the normal vector of mode 𝑗 /5/.

3.7 Inertial coefficients 𝑀𝑗𝑘

The inertial coefficient matrix 𝑀𝑗𝑘 is set up under the assumption of free stable floatation
with no external constraints. The mass is expressed as

© DHI - Frequency Response Calculator and Mooring Analysis 13

Basic Equations

𝑚 = 𝜌∀

𝑥𝑏 = 𝑥𝑔 , 𝑦𝑏 = 𝑦𝑔

𝑚 0 0 0 𝑚𝑧𝑔 −𝑚𝑦𝑔
0 𝑚 0 −𝑚𝑧𝑔 0 𝑚𝑥𝑔 (3.52)

0 0 𝑚 𝑚𝑦𝑔 −𝑚𝑥𝑔 0
𝑀=
0 −𝑚𝑧𝑔 𝑚𝑦𝑔 𝐼11 𝐼12 𝐼13
𝑚𝑧𝑔 0 −𝑚𝑥𝑔 𝐼21 𝐼22 𝐼23
[−𝑚𝑦𝑔 𝑚𝑥𝑔 0 𝐼31 𝐼32 𝐼33 ]

and the horizontal centre of gravity coordinates, denoted with index 𝑔, are set as the
centre of buoyancy, denoted with index 𝑏 (see Section 3.1.2). The vertical location of the
centre of gravity depends on the weight distribution of the floating body (usually a ship),
and largely depends on whether it is loaded or unloaded.

The moments of inertia

𝐼𝑖𝑗 = 𝜌∀𝑟𝑖𝑗 |𝑟𝑖𝑗 | (3.53)

are obtained from the radii of gyration. The radii of gyration input consist of a (3×3) array
of values of 𝑟 /5/.

3.8 External force 𝐹𝑗,𝑛𝑙 (𝑡)

The total external force

𝐹𝑗,𝑛𝑙 (𝑡) = 𝐹𝑗,𝑚𝑜𝑜𝑟 (𝑡) + 𝐹𝑗,𝑣𝑖𝑠𝑐 (𝑡) + 𝐹𝑗,𝑤 (𝑡) + 𝐹𝑗,𝑐 (𝑡) + 𝐹𝑗,𝑑𝑟𝑖𝑓𝑡 (𝑡)
(3.54)

is the sum of the mooring forces 𝐹𝑗,𝑚𝑜𝑜𝑟 (𝑡), viscous damping forces 𝐹𝑗,𝑣𝑖𝑠𝑐 (𝑡), wind
forces 𝐹𝑗,𝑤𝑖𝑛𝑑 (𝑡), current forces 𝐹𝑗,𝑐𝑢𝑟 (𝑡), slow wave drift forces 𝐹𝑗,𝑑𝑟𝑖𝑓𝑡 (𝑡) and frictional
damping forces 𝐹𝑗,𝑓𝑟𝑖𝑐 (𝑡) in the mooring system in each degree of freedom 𝑗. They are
described in this section.

3.8.1 Viscous damping forces 𝐹𝑗,𝑣𝑖𝑠𝑐 (𝑡)

The viscous damping force is defined as

𝐹𝑗,𝑣𝑖𝑠𝑐 (𝑡) = 𝐵𝑗0 1

+ ∑[𝐵𝑗𝑘 2
𝑥̇ 𝑘 (𝑡) + 𝐵𝑗𝑘 3 3 (𝑡)]
𝑥̇ 𝑘 (𝑡)|𝑥̇ 𝑘 (𝑡)| + 𝐵𝑗𝑘 𝑥̇ 𝑘 (3.55)
𝑘=1

in which 𝐵0 , 𝐵1 , 𝐵2 and 𝐵3 are the constant, linear, quadratic and cubic damping
coefficient matrices. 𝑥̇ 𝑘 (𝑡) is the speed of motion into direction 𝑘 .

© DHI - Frequency Response Calculator and Mooring Analysis 14

Basic Equations

3.8.2 Mooring forces 𝐹𝑗,𝑚𝑜𝑜𝑟 (𝑡)

Mooring lines
Given that the displacement of a floating body is expressed as

𝑑𝑙 = 𝑙 − 𝑙0 (3.56)

with the line length 𝑙 and the unstretched line length 𝑙0 , then the elongation is proportional
to the force in the mooring line 𝑑𝑙~𝐹𝑗,𝑚𝑜𝑜𝑟,𝑙𝑖𝑛𝑒 (𝑡). This relation is provided by means of a
Load-excursion curve denoted as 𝑅(𝑑𝑙),

𝐹𝑗,𝑚𝑜𝑜𝑟,𝑙𝑖𝑛𝑒 (𝑡) = 𝑅(𝑑𝑙) + 𝐹𝑗,𝑙𝑖𝑛𝑒,𝑑𝑎𝑚𝑝 (𝑡)

(3.57)

which is line-material specific and which is provided as input data. From this relation, the
mooring line force is calculated. 𝐹𝑙𝑖𝑛𝑒,𝑑𝑎𝑚𝑝 (𝑡) is a damping force, which can be applied
for the purpose of calibration. It is governed by linear and quadratic damping coefficients
𝐵1 and 𝐵2 as part of the input data. It is derived with the relation
6
1 2
𝐹𝑗,𝑙𝑖𝑛𝑒,𝑑𝑎𝑚𝑝 (𝑡) = ∑[𝐵𝑗𝑘 𝑥̇ 𝑘 (𝑡) + 𝐵𝑗𝑘 𝑥̇ 𝑘 (𝑡)|𝑥̇ 𝑘 (𝑡)|] ∙ 𝑙𝑖𝑛𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑖𝑜𝑛𝑗 (𝑡) (3.58)
𝑘=1

where 𝑥̇ 𝑘 (𝑡) is the speed of motion into direction 𝑘 . 𝑙𝑖𝑛𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑖𝑜𝑛𝑗 (𝑡) is a vector
pointing into the direction of the line.

Fenders
If the fender attachment point moves along the fender force direction with a displacement
𝑑𝑙 , then the fender is compressed by the same length. The fender compression is related
to the fender force 𝑑𝑙~𝐹𝑛 (𝑡) by means of the fender deflection-reaction curve

𝐹𝑗,𝑛 (𝑡) = 𝑅(𝑑𝑙) + 𝐹𝑗,𝑓𝑒𝑛𝑑𝑒𝑟,𝑑𝑎𝑚𝑝 (𝑡)

(3.59)

which is fender-specific and which is provided as input data. The fender damping force

6
1
𝐹𝑗,𝑓𝑒𝑛𝑑𝑒𝑟,𝑑𝑎𝑚𝑝 (𝑡) = ∑ 𝐵𝑗𝑘 𝑥̇ 𝑘 (𝑡) ∙ 𝑓𝑒𝑛𝑑𝑒𝑟 𝑓𝑜𝑟𝑐𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑖𝑜𝑛𝑗 (𝑡) (3.60)
𝑘=1

in which 𝑥̇ 𝑘 (𝑡) is the speed of motion into direction 𝑘 and 𝐵1 is the linear damping
coefficient. 𝑓𝑒𝑛𝑑𝑒𝑟 𝑓𝑜𝑟𝑐𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑖𝑜𝑛𝑗 (𝑡) is a vector pointing into the direction of the
fender force.

Furthermore, the fender friction is calculated by

𝑣𝑒𝑙𝑗 (𝑡)
𝐹𝑗,𝑓𝑟𝑖𝑐 (𝑡) = 𝜇𝐹𝑗,𝑛 (𝑡) ∙
|𝑣𝑒𝑙(𝑡)| (3.61)

© DHI - Frequency Response Calculator and Mooring Analysis 15

Basic Equations

with the fender friction coefficient 𝜇 and the relative velocity between ship attachment
point and fender attachment point 𝑣𝑒𝑙(𝑡). The fender force then reads

𝐹𝑗,𝑚𝑜𝑜𝑟,𝑓𝑒𝑛𝑑𝑒𝑟 (𝑡) = 𝐹𝑗,𝑛 (𝑡) + 𝐹𝑗,𝑓𝑟𝑖𝑐 (𝑡)

(3.62)

Chains
The chain tension is derived using the catenary equation

𝑑2 𝑧 𝑑𝑧 2
𝐻(𝑡) = −𝑚𝑔 √1 + ( ) (3.63)
𝑑𝑥 2 𝑑𝑥

In which 𝑚𝑔 is the (submerged) chain weight, 𝐻(𝑡) is the horizontal force of the chain at
𝑑𝑥
rest, and is the local orientation of the chain element 𝑑𝑠 /6/.
𝑑𝑧

The equation is solved for 𝐻 using the Newton-Raphson method, in which 𝐻 is used as
initial estimate, which finally leads to 𝐹𝑗,𝑚𝑜𝑜𝑟,𝑐ℎ𝑎𝑖𝑛 . In this method, it is considered that the
maximum peak tension is

𝑇𝑚𝑎𝑥 = 𝐻 + 𝑚𝑔𝑑 (3.64)

with the sag of the catenary 𝑑 . It is further defined that the cable length

2𝐻
𝐿0 = √ +1
𝑚𝑔𝑑0 (3.65)

and the horizontal distance from the touch-down point to the surface point at rest is

𝐻 𝑑0 𝑚𝑔
𝑥0 = cosh−1 ( + 1)
𝑚𝑔 𝐻 (3.66)

3.8.3 Wind forces

Wind forces are derived based on the OCIMF method /9/, in which

1
𝐹𝑥,𝑤 (𝑡) = 𝐶 𝜌 𝑉 (𝑡)2 𝐴 𝑇,𝑤
2 𝑥,𝑤 𝑎𝑖𝑟 𝑤
1
𝐹𝑦,𝑤 (𝑡) = 𝐶 𝜌 𝑉 (𝑡)2 𝐴𝐿,𝑤
2 𝑦,𝑤 𝑎𝑖𝑟 𝑤 (3.67)

1
𝑀𝑥𝑦,𝑤 (𝑡) = 𝐶 𝜌 𝑉 (𝑡)2 𝐴𝐿,𝑤 𝐿𝐵𝑃
2 𝑥𝑦,𝑤 𝑎𝑖𝑟 𝑤

© DHI - Frequency Response Calculator and Mooring Analysis 16

Basic Equations

with the density of air 𝜌𝑎𝑖𝑟 , the transversal windage area 𝐴 𝑇,𝑤 , the longitudinal windage
area 𝐴𝐿,𝑤 , the length between perpendiculars 𝐿𝐵𝑃 , the wind speed 𝑉𝑤 (𝑡). The wind drag
coefficients 𝐶𝑥,𝑤 , 𝐶𝑦,𝑤 and 𝐶𝑥𝑦,𝑤 depend on the vessel type, the loading condition of the
vessel and the angle of wind attack, e.g. see /3/ and /9/.

3.8.4 Current forces

Current forces are derived based on the OCIMF method /9/, in which

1
𝐹𝑥,𝑐 (𝑡) = 𝐶 𝜌 𝑉 (𝑡)2 𝐿𝐵𝑃 𝑇
2 𝑥,𝑐 𝑤𝑎𝑡𝑒𝑟 𝑐
1
𝐹𝑦,𝑐 (𝑡) = 𝐶 𝜌 𝑉 (𝑡)2 𝐿𝐵𝑃 𝑇
2 𝑦,𝑐 𝑤𝑎𝑡𝑒𝑟 𝑐 (3.68)

1
𝑀𝑥𝑦,𝑐 (𝑡) = 𝐶 𝜌 𝑉 (𝑡)2 𝐿𝐵𝑃 𝑇
2 𝑥𝑦,𝑐 𝑤𝑎𝑡𝑒𝑟 𝑐

with the density of water 𝜌𝑤𝑎𝑡𝑒𝑟 , the transversal area affected by the current 𝐴 𝑇,𝑐 , the
longitudinal area affected by the current 𝐴𝐿,𝑐 , the length between perpendiculars 𝐿𝐵𝑃 , the
current speed 𝑉𝑐 . The areas affected by the current are determined from the vessel hull.
The current drag coefficients 𝐶𝑥,𝑐 , 𝐶𝑦,𝑐 and 𝐶𝑥𝑦,𝑐 depend on the water depth to draft ratio
of the vessel and the angle of current attack, e.g. see /9/. The OCIMF method uses a
time series of current speed and direction.

However, this method cannot produce the drag-induced yaw moment under shear current
flow conditions. Therefore, the panel method has been devised in order to calculate the
current drag-induced yaw moment on the vessel for shear current. The input for this
method is a two-dimensional (2D) current field consisting of the parameters water depth,
and fluxes in 𝑥 and 𝑦 direction. The method involves

• Calculation of the current velocities in each element of the 2D flow field,

• Interpolation of these velocities onto each panel
• Calculation of the average current speeds in 𝑥 and 𝑦 direction
• Calculation of the longitudinal and lateral drag force using the OCIMF method /9/.
• Projection of the current vectors onto the vessel grid panels
• Calculation of the force in each panel from the yaw moment using the drag equation
(see equation (3.68)).

© DHI - Frequency Response Calculator and Mooring Analysis 17

Numerical Implementation

4 Numerical Implementation
In this section, the fundamental solution algorithms are given, and additional numerical
aspects are commented. Figure 4.1 replicates the flow chart variables from Section 3, but
additionally shows the distribution of the flow between the modules Frequency Response
Calculator (FRC) and MIKE 21 Mooring Analysis (MA).

FRC MIKE 21 MA

Figure 4.1 Flow chart of variables. The frames indicate the modules, dealing with the variables.

In Section 4.1, the implementation of the theory is provided. Section 4.2 outlines the
treatment of the irregular frequencies, which can occur during the application of the
boundary element method (BEM). Finally, the difference between Final computation
mode and Convergence mode in MIKE 21 MA is explained.

4.1 Implementation of the theory

MIKE 21 MA solves the time-domain equations of motion (3.6) to obtain a time history of
body motion in six degrees of freedom, due to a specified incident wave field, and
mooring arrangement. This integration is performed using the explicit Runge-Kutta 4th
order scheme, and a direct inversion of the inertia matrix.

The required hydrodynamic, hydrostatic and inertia coefficients of the body are computed
using the low-order panel method program, FRC. FRC is used for performing a complete
radiation analysis of the structure, that is 𝐴𝑗𝑘 (𝜔), 𝐵𝑗𝑘 (𝜔), and 𝜑𝑗 (𝜔) are computed at
evenly spaced frequencies over the entire significant domain of frequencies (including
𝜔 = 0, ∞).

We then use the fact that 𝑎𝑗𝑘 = 𝐴𝑗𝑘 (∞), and compute 𝐾𝑗𝑘 from the damping coefficients
via equation (3.22). The inertia matrix, 𝑀𝑗𝑘 , and the hydrostatic matrix, 𝐶𝑗𝑘 , are direct
output from FRC.

© DHI - Frequency Response Calculator and Mooring Analysis 18

Numerical Implementation

The incident wave exciting force, 𝐹𝑗𝐷 (𝑡), can be defined by providing the incident wave
pressure and velocity at each panel centroid on the body (or these data can be computed
by linearising the results of a MIKE 21 BW calculation over a rectangle of the free surface
enclosing the body). In this case, the Haskind relations and the Fourier transform (3.50)
are used for combining the incident wave data with the FRC radiation potentials to get the
exciting forces. It can be shown that by using the complex conjugate of the radiation
potentials in the Haskind relations, the correct forces are found. All Fourier transforms are
done using the Fast Fourier Transformation (FFT).

Finally, instantaneous (non-linear) point forces on the structure due to each mooring line
(or fender, or post) are computed based on the specified properties of the line/fender/post
and the relative positions of the attachment points. These forces as well as the viscous
damping forces proportional to a power of the body velocity, are all included at each sub-
time step of the time integration.

4.2 Irregular frequencies

When a surface piercing body is analysed by boundary element techniques, numerical
errors can occur at a number of irregular frequencies. In general, these irregular
frequencies are higher than the frequency range of practical interest. DHI has
implemented a technique, which first predicts the location of the irregular frequencies and
smooth’s the velocity potential around these frequencies using iterative smoothing
techniques (see /12/). This method effectively minimises the irregular frequency effect
while still allowing for a fast run time.

The location of irregular frequencies can be calculated analytically for simple geometries.
The analytical calculations for a rectangular prism can be extended through an equivalent
box formulation to allow for the prediction of irregular frequencies 𝜔𝑝𝑚 in an arbitrarily
shaped 3D structure. This procedure is demonstrated in equation (4.1).

𝐴𝑤 𝐴𝑚 𝐴𝑐
𝐶𝑤 = , 𝐶𝑚 = , 𝐶𝑐 =
𝐿𝑂𝐴∗𝐵𝑒𝑎𝑚 𝐵𝑒𝑎𝑚∗𝐷𝑟𝑎𝑓𝑡 𝐿𝑂𝐴∗𝐷𝑟𝑎𝑓𝑡

𝐿𝑂𝐴−𝐵𝑒𝑎𝑚 𝐿𝑂𝐴−𝐵𝑒𝑎𝑚
3 1+6∗ ∗ln 𝑝 1+6∗ ∗ln 𝑚
𝐿𝑂𝐴+𝐵𝑒𝑎𝑚 𝐿𝑂𝐴+𝐵𝑒𝑎𝑚
𝐶0 = , 𝐶1 = , 𝐶2 =
4 8 8

𝐵0 = 𝐶𝑤 𝐶0 , 𝐵1 = 𝐶𝑐 𝐶1 , 𝐵2 = 𝐶𝑚 𝐶2

𝐿𝑂𝐴2 𝐵𝑒𝑎𝑚2
𝛼1 = , 𝛼2 =
𝐿𝑂𝐴2 +𝐵𝑒𝑎𝑚2 𝐿𝑂𝐴2 +𝐵𝑒𝑎𝑚2
(4.1)
𝛼 𝛼 ∀
𝐿𝑒 = 𝐵1 ∗ 𝐵0 1 , 𝐵𝑒 = 𝐵2 ∗ 𝐵0 2 , ℎ𝑒 =
𝐿𝑒 ∗𝐵𝑒

𝑝 2 𝑚2
𝑘𝑝𝑚 = 𝜋 ∗ √( + )
𝐿𝑒 𝐵𝑒

1
𝜔𝑝𝑚 = √𝑔𝑘𝑝𝑚 ∗
tanh(𝑘𝑝𝑚 ℎ𝑒 )

where 𝐴𝑤 , 𝐴𝑚 and 𝐴𝑐 are the water plane, midsection and central longitudinal plane
areas, and 𝐿𝑂𝐴 is the length over all the vessel. 𝑝 and 𝑚 are indices from 1 to infinity,
and 𝑘 is the wave number.

© DHI - Frequency Response Calculator and Mooring Analysis 19

Numerical Implementation

4.3 Convergence mode

A simulation in convergence mode is a full simulation, in which usually no external forces
are applied (although the user can apply long-term conditions such as river currents).
Starting with an initial displacement of 0 in all six degrees of freedom, the simulation is
repeated, until
1. the displacement at the end of the simulation does not exceed a threshold, given by
the user OR
2. the maximum number of simulations (iterations) is reached.

The purpose of the convergence model is to achieve an equilibrium location of the vessel
under the given mooring system prior to initiating the actual mooring simulation with e.g.
wave forcing. When setting up the mooring system and pre-tensions, the system may not
be perfectly balanced. This is ensured in the convergence model simulation.

© DHI - Frequency Response Calculator and Mooring Analysis 20

Output Data

5 Output Data
The output data of both Frequency Response Calculator (FRC) and MIKE 21 Mooring
Analysis (MA) modules are summarised in this section.

5.1 Frequency Response Calculator

FRC provides the following results:

• Global output files

- Vessel response file
- Added mass coefficients
- Damping coefficients
- Drift forces
• Vessel specific output files
- Response amplitude operator
- Exciting forces

5.1.1 Global output files

The vessel response file summarises all hydrostatic, gyrostatic and geometry data of the
investigated floating body system such as

• Water depth
• Water density
• Vessel shape
• Vertical centre of gravity
• Draft radii of gyration
• Length over all
• Beam
• Maximum vessel height
• Submerged volume
• Added mass coefficients
• Damping coefficients
• Wave diffraction coefficients

The file is used as input for the subsequent calculations in MIKE 21 MA.

The added mass coefficients result file contains the added mass response for each
degree of freedom as a function of wave frequency. The added mass is contained in a
6x6 matrix for each wave frequency and can be described as an inertia added to the
system due to an acceleration of the vessel.

The damping coefficients result file contains the radiation damping of the vessel and is in
the same format as the added mass. Radiation damping describes the energy lost in the
vessel as it creates waves through motion.

The drift forces result file contains the second order drift surge and sway forces and yaw
moments for each frequency and direction specified. This output file can only be
generated if wave drift forces are included in the calculation.

© DHI - Frequency Response Calculator and Mooring Analysis 21

Output Data

5.1.2 Vessel-specific output files

The response amplitude operators result file contains the amplitude and phase data of
the response amplitude operator (RAO) for each frequency and direction specified. The
RAO is a transfer function that can be used to determine the effect that a sea state will
have upon the motion of a vessel.

The exciting forces result file contains the amplitude and phase data of the exciting forces
and moments for each frequency and direction specified. The exciting force is the first
order wave force on the vessel and is calculated using the Haskind relation (see
Appendix A). The exciting force describes the combination of the incident wave forces
from a harmonic wave and the diffraction force due to the vessel on the undisturbed wave
field.

5.2 MIKE 21 MA
MIKE 21 MA provides the following results in both simulation modes:

• Global output files

- Line forces
- Fender forces
- Line pre-tension deviation (in convergence mode only)
• Vessel specific output files
- Vessel motion
- Mooring force on vessel
- External force
- Surface elevation (in final computation mode only)
- Chain forces

5.2.1 Global output files

The line forces result file contains an item for every specified mooring line, showing the
dynamically computed line force at each time step.

This fender forces result file contains an item for every specified spatial fender, showing
the dynamically computed fender force as a function of time.

In the convergence mode (see Section 4.3), the line pre-tension deviation is provided. It
reflects the success of the convergence iteration process. This file contains an item for
every specified spatial mooring line, and all these items will, at each iteration, contain
data values representing the percentage deviation between the dynamically computed
line tension (at the iteration) and the line pre-tension. If a convergence is successful, then
this file will at the last iteration step contain values, which are all smaller or equal the Line
convergence threshold on the Convergence parameters dialog.

5.2.2 Vessel-specific output files

The vessel motion result file contains six items for the six degrees of freedom for each
vessel, as a function of time. The definition of the items is provided in Figure 3.2.

The mooring force on vessel result file contains six items (three force items, three
direction items) representing the force from mooring system on the vessel for each time
step.

© DHI - Frequency Response Calculator and Mooring Analysis 22

Output Data

The external forces result file contains six items (three force items, three direction items)
for each external forcing applied (wind, current, 2nd order drift forces), thus representing
the force from each external forcing on the vessel for each time step.

The surface elevation result file contains the surface elevation at the vessels midpoint for
each time step.

The diffraction force result file contains six items (three force items, three direction items)
representing the diffraction force on the vessel for each time step.

The chain forces result file contains as many items as there are chains declared. The file
displays the chain force evolution for each time step.

© DHI - Frequency Response Calculator and Mooring Analysis 23

References

6 References
/1/ Babarit, A., Delhommeau, G. (2015): “Theoretical and numerical aspects of the
open source BEM solver NEMOH”, LHEEA CNRS UMR 6598, Ecole Centrale
de Nantes, France.

/2/ Bingham, H.B. (2000): “A hybrid Boussinesq-panel method for predicting the
motion of a moored ship”, Coastal Engineering, 40, 21-38.

/3/ Brix, J. (1993): “Manoeuvring Technical Manual”, Seehafen Verlag, Hamburg,

Germany.

/4/ Delhommeau, G. (1987): “Les Problèmes de Diffraction-Radiation et de

Résistance de Vagues: Etude Théorique et Résolution Numérique par la
Méthode des Singularites”, PhD thesis, Ecole Nationale Supérieure de
Mécanique, Laboratoire d’Hydrodynamique Navale, Nantes, France.

/5/ Delhommeau, G. (1993): “Seakeeping Codes AQUADYN and AQUAPLUS”, In

Proc. of the 19tzhWegemt School on Numerical Simulation of Hydrodynamics:
Ships and Offshore Structures, Nantes, France.

/6/ Irvine, H.M. (2000): “Cable Structures”, MIT Press, Cambridge, MA, USA.

/7/ Lighthill, J. (1979): “Waves and hydrodynamic loading”, Proc. of the 2nd Int.
Conf. on the Behaviour of Offshore Structures, BOSS ’79, Vol. 1, 1-40.

/8/ Newman, J.N. (1977): “Marine Hydrodynamics”, MIT Press, Cambridge,

Massachusetts, USA.

/9/ Oil Companies International Marine Forum (2008): “Mooring Equipment

Guidelines”, 3rd edition, Witherby Seamanship International, Livingston, UK.

/10/ Parisella, G., Gourlay, T.P. (2016): “Comparison of open-source code Nemoh
with Wamit for cargo ship motions in shallow water”, Research report 2016-23,
Centre for Marine Science and Technology, Curtin University, October 2016,
Perth, WA, Australia.

/11/ Wehausen, J.V. and Laitone, E.V. (1960): “Surface Waves in Fluid Dynamics
III”, in Handbuch der Physik 9, Eds.: S. Fluegge and C. Truesdell, pp 446-778,
Springer Verlag.

/12/ Zhu, X. and Lee, C.H. (1994): "Removing the Irregular Frequencies in Wave-
Body Interactions", 9th International Workshop on Water Waves and Floating
Bodies, Kizu, Japan.

/13/ Delhommeau, G. (1993). Numerical Simulation of Hydrodynamics: Ships and

Offshore Structures. Nineteenth WEGMENT School.

/14/ Newman, J.N. (1974). Second-order Slowly-varying Forces on Vessels in

Irregular Waves. In Int. Symposium on Dynamics of Marine Vehicles and
Structures in waves.

6.1 Further reading on MIKE 21 MA and its predecessors

• Bhautoo, P. et. al. (2015): “Moored vessel interaction induced by passing ships at
the Port of Brisbane”, Australasian Coasts & Ports Conference, 15-18 September
2015, Auckland, New Zealand.

© DHI - Frequency Response Calculator and Mooring Analysis 24

References

• Brüning, A.et. al. (2011): “A hybrid modelling approach for floating breakwater
dimensioning”, 5th International Short Conference on Applied Coastal Research
2011.
• Brüning, A. et. al. (2014): “Assessing mooring forces at an offshore wind terminal in
Bremerhaven, Germany”, PIANC World Congress, San Francisco, USA.
• Hansen, H.F. et al. (2009): “Multi vessel interaction in shallow water”, Proceedings of
the 28th International Conference on Ocean, Offshore and Arctic Engineering
OMAE2009, Honolulu, Hawaii, USA, May 31 - June 5, 2009.
• Hansen, H.F., Kirkegaard, J and Kofoed-Hansen, H. (2010): “Hydraulic aspects of
marine LNG receiving terminal design”, Rio Oil & Gas Expo and Conference, Rio de
Janeiro, September 13-16, 2010.
• Harkin, A., Mortensen, S. and Dixen, M. (2017): “Validation of Moored Vessel
Response Simulator with Physical Model Comparisons”, Coasts & Ports 2017
Conference – Cairns, 21-23 June 2017.
• Jensen, B., Hansen, H.F. and Kirkegaard, J. (2015): “Estimating Quadratic Transfer
Functions for Floating Structures using Model Test Data from Irregular Sea States”,
Proceedings of the Twenty-fifth (2015) International Ocean and Polar Engineering
Conference, Kona, Big Island, Hawaii, USA, June 21-26, 2015
• Kerper, D. et. al. (2017): “Nearshore FSO Response under Rapidly Varying
Landslide Tsunami Hydrodynamic Loads”, Proceedings of the Twenty-seventh
(2017) International Ocean and Polar Engineering Conference, San Francisco, CA,
USA, June 25-30, 2017.
• Mortensen, S, Allery, C, Kirkegaard, J and Hancock, R, (2008): “Numerical Modelling
of Moored Vessel Motions Caused by Passing Vessels in the Port of Brisbane”,
Proceedings of the ASME 27th International Conference on Offshore Mechanics and
Arctic Engineering OMAE 2008, June 15-20, 2008, Estoril, Portugal.
• Mortensen, S, Allery, C, Kirkegaard, J and Hancock, R. (2009): “Numerical Modelling
of Moored Vessel Motions Caused by Passing Vessels”, Coast & Ports 2009
Conference Proceedings, Wellington, New Zealand.

© DHI - Frequency Response Calculator and Mooring Analysis 25

 APPENDIX A
The Haskind Relation

© DHI - Frequency Response Calculator and Mooring Analysis

The Haskind Relation

A The Haskind Relation

The Haskind relation is a way of finding the diffraction force without estimating the
scattered potential. Instead, the force is found in a combination of the incoming wave field
and the potential for the radiated wave field, which is the wave potential found for a
moving body in a calm sea.

Expressed in a simplified way, the diffraction force is found by multiplying the radiated
body movement on the incoming wave pressure field minus the radiation pressure field
multiplied by the velocity of the incoming wave field. This seems quite abstract; therefore,
the following heuristic explanation is given.

We have to go to structural engineering, where this principle is used extensively. In this

field, the above statement can be found from the application of Betti’s equation, also
known as the reciprocity principle. In this case, a reduced form of the principle says

𝐹 𝐴 𝑢𝐵 = 𝐹 𝐵 𝑢 𝐴 (A.1)

In words, the equation says that for a system, a force from case A multiplied by the
movement of case B is equal to the force from case B multiplied by the movement of
case A. This equation is given without further proof; however, it is a well-established
technique in structural engineering.

The diffraction force on the body can be written as the sum of a contribution from the
incoming wave field and the scattered wave field:

𝐹𝐷 = 𝐹𝐼 + 𝐹𝑆 (A.2)

where 𝐹𝐷 is the total diffraction force, 𝐹𝐼 is the force due to the incoming wave field, and
𝐹𝑆 is the force due to the scattered wave field. Note that 𝐹𝐼 is also known as Froude-
Krylov force. In the equation, there are two unknowns 𝐹𝐷 that we want to find and 𝐹𝑆 that
has to be estimated. Let us say that the radiation potential is known and therefore also
the force, 𝐹𝑅 , using Betti’s equation (A.1)

𝐹𝑅 𝑢𝑆 = 𝐹𝑆 𝑢𝑅 (A.3)

In the above equation, 𝑢𝑆 is apparently unknown, but at the body surface it can be related
to the velocity found from the incoming wave, 𝑢𝑆 = −𝑢𝐼 . Then, equation (A.2) is
multiplied by 𝑢𝑅 . Using equation (A.3) and substituting 𝑢𝑆 with −𝑢𝐼 , gives the Haskind
relation

𝐹𝐷 𝑢𝑅 = 𝐹𝐼 𝑢𝑅 − 𝐹𝑅 𝑢𝐼 (A.4)

This is basically the Haskind relation in a schematised form. The approach followed here
is similar to the approach described in /7/ where it was used to find the second order
force components. The derivation of the Haskind relations from conventional methods is
given in /8/.

© DHI - Frequency Response Calculator and Mooring Analysis A-1