MARINE

DIVISION

3058
716A

Starspec

User guide

Issue 0 1 2
Date Mars 2010 July 2013 July 2019
Author Guillaume de Guillaume de Guillaume de
Hauteclocque Hauteclocque Hauteclocque
Verified by Flávia Rezende Charles Monroy Charles Monroy
Comments StarSpec v1.09 StarSpec v1.3 StarSpec v8.1

Bureau Veritas Research Department - Immeuble le ’1828’, 67/71 Bd du Château - 92571 Neuilly sur Seine -France
Tel: +33 1 55 24 70 00 / www.bureauveritas.com
 MARINE DIVISION – GENERAL CONDITIONS

ARTICLE 1 5.4. - The Services of the Society cannot create any obligation bearing on the Society or
1.1. - BUREAU VERITAS is a Society the purpose of whose Marine Division (the “Society”) constitute any warranty of proper operation, beyond any representation set forth in the
is the classification (“Classification”) of any ship or vessel or structure of any type or part of Rules, of any Unit, equipment or machinery, computer software of any sort or other
it or system therein collectively hereinafter referred to as a “Unit” whether linked to shore, comparable concepts that has been subject to any survey by the Society.
river bed or sea bed or not, whether operated or located at sea or in inland waters or partly
on land, including submarines, hovercrafts, drilling rigs, offshore installations of any type ARTICLE 6
and of any purpose, their related and ancillary equipment, subsea or not, such as well head 6.1. - The Society accepts no responsibility for the use of information related to its Services
and pipelines, mooring legs and mooring points or otherwise as decided by the Society. which was not provided for the purpose by the Society or with its assistance.
The Society: 6.2. - If the Services of the Society cause to the Client a damage which is proved to
• prepares and publishes Rules for classification, Guidance Notes and other documents be the direct and reasonably foreseeable consequence of an error or omission of
(“Rules”); the Society, its liability towards the Client is limited to ten times the amount of fee
• issues Certificates, Attestations and Reports following its interventions (“Certificates”); paid for the Service having caused the damage, provided however that this limit
• publishes Registers. shall be subject to a minimum of eight thousand (8,000) Euro, and to a maximum
1.2. - The Society also participates in the application of National and International which is the greater of eight hundred thousand (800,000) Euro and one and a half
Regulations or Standards, in particular by delegation from different Governments. Those times the above mentioned fee.
activities are hereafter collectively referred to as “Certification”. The Society bears no liability for indirect or consequential loss such as e.g. loss of
1.3. - The Society can also provide services related to Classification and Certification such revenue, loss of profit, loss of production, loss relative to other contracts and
as ship and company safety management certification; ship and port security certification, indemnities for termination of other agreements.
training activities; all activities and duties incidental thereto such as documentation on any 6.3. - All claims are to be presented to the Society in writing within three months of the date
supporting means, software, instrumentation, measurements, tests and trials on board. when the Services were supplied or (if later) the date when the events which are relied on of
1.4. - The interventions mentioned in 1.1., 1.2. and 1.3. are referred to as “Services”. The were first known to the Client, and any claim which is not so presented shall be deemed
party and/or its representative requesting the services is hereinafter referred to as the waived and absolutely barred.
“Client”. The Services are prepared and carried out on the assumption that the
Clients are aware of the International Maritime and/or Offshore Industry (the ARTICLE 7
“Industry”) practices. 7.1. - Requests for Services are to be in writing.
1.5. - The Society is neither and may not be considered as an Underwriter, Broker in ship’s 7.2. - Either the Client or the Society can terminate as of right the requested
sale or chartering, Expert in Unit’s valuation, Consulting Engineer, Controller, Naval Services after giving the other party thirty days' written notice, for convenience, and
Architect, Manufacturer, Shipbuilder, Repair yard, Charterer or Shipowner who are not without prejudice to the provisions in Article 8 hereunder.
relieved of any of their expressed or implied obligations by the interventions of the Society. 7.3. - The class granted to the concerned Units and the previously issued certificates
remain valid until the date of effect of the notice issued according to 7.2. hereabove subject
ARTICLE 2 to compliance with 2.3. hereabove and Article 8 hereunder.
2.1. - Classification is the appraisement given by the Society for its Client, at a certain date,
following surveys by its Surveyors along the lines specified in Articles 3 and 4 hereafter on ARTICLE 8
the level of compliance of a Unit to its Rules or part of them. This appraisement is 8.1. - The Services of the Society, whether completed or not, involve the payment of fee
represented by a class entered on the Certificates and periodically transcribed in the upon receipt of the invoice and the reimbursement of the expenses incurred.
Society’s Register. 8.2. - Overdue amounts are increased as of right by interest in accordance with the
2.2. - Certification is carried out by the Society along the same lines as set out in Articles 3 applicable legislation.
and 4 hereafter and with reference to the applicable National and International Regulations 8.3. - The class of a Unit may be suspended in the event of non-payment of fee after
or Standards. a first unfruitful notification to pay.
2.3. - It is incumbent upon the Client to maintain the condition of the Unit after
surveys, to present the Unit for surveys and to inform the Society without delay of ARTICLE 9
circumstances which may affect the given appraisement or cause to modify its 9.1. - The documents and data provided to or prepared by the Society for its Services, and
scope. the information available to the Society, are treated as confidential. However:
2.4. - The Client is to give to the Society all access and information necessary for the • Clients have access to the data they have provided to the Society and, during the period
performance of the requested Services. of classification of the Unit for them, to the classification file consisting of survey
reports and certificates which have been prepared at any time by the Society for the
ARTICLE 3 classification of the Unit ;
3.1. - The Rules, procedures and instructions of the Society take into account at the • copy of the documents made available for the classification of the Unit and of available
date of their preparation the state of currently available and proven technical survey reports can be handed over to another Classification Society Member of the
knowledge of the Industry. They are not a code of construction neither a guide for International Association of Classification Societies (IACS) in case of the Unit’s transfer
maintenance or a safety handbook. of class;
Committees consisting of personalities from the Industry contribute to the development of • the data relative to the evolution of the Register, to the class suspension and to the
those documents. survey status of the Units are passed on to IACS according to the association working
3.2. - The Society only is qualified to apply its Rules and to interpret them. Any rules;
reference to them has no effect unless it involves the Society’s intervention. • the certificates, documents and information relative to the Units classed with the Society
3.3. - The Services of the Society are carried out by professional Surveyors according to may be reviewed during IACS audits and are disclosed upon order of the concerned
the Code of Ethics of the Members of the International Association of Classification governmental or inter-governmental authorities or of a Court having jurisdiction.
Societies (IACS). The documents and data are subject to a file management plan.
3.4. - The operations of the Society in providing its Services are exclusively
conducted by way of random inspections and do not in any circumstances involve ARTICLE 10
monitoring or exhaustive verification. 10.1. - Any delay or shortcoming in the performance of its Services by the Society arising
from an event not reasonably foreseeable by or beyond the control of the Society shall be
ARTICLE 4 deemed not to be a breach of contract.
4.1. - The Society, acting by reference to its Rules:
• reviews the construction arrangements of the Units as shown on the documents ARTICLE 11
presented by the Client; 11.1. - In case of diverging opinions during surveys between the Client and the Society’s
• conducts surveys at the place of their construction; surveyor, the Society may designate another of its surveyors at the request of the Client.
• classes Units and enters their class in its Register; 11.2. - Disagreements of a technical nature between the Client and the Society can be
• surveys periodically the Units in service to note that the requirements for the submitted by the Society to the advice of its Marine Advisory Committee.
maintenance of class are met.
The Client is to inform the Society without delay of circumstances which may cause ARTICLE 12
the date or the extent of the surveys to be changed. 12.1. - Disputes over the Services carried out by delegation of Governments are assessed
within the framework of the applicable agreements with the States, international
ARTICLE 5 Conventions and national rules.
5.1. - The Society acts as a provider of services. This cannot be construed as an 12.2. - Disputes arising out of the payment of the Society’s invoices by the Client are
obligation bearing on the Society to obtain a result or as a warranty. submitted to the Court of Nanterre, France.
5.2. - The certificates issued by the Society pursuant to 5.1. here above are a 12.3. - Other disputes over the present General Conditions or over the Services of
statement on the level of compliance of the Unit to its Rules or to the documents of the Society are exclusively submitted to arbitration, by three arbitrators, in London
reference for the Services provided for. according to the Arbitration Act 1996 or any statutory modification or re-enactment
In particular, the Society does not engage in any work relating to the design, thereof. The contract between the Society and the Client shall be governed by
building, production or repair checks, neither in the operation of the Units or in their English law.
trade, neither in any advisory services, and cannot be held liable on those accounts.
Its certificates cannot be construed as an implied or express warranty of safety, ARTICLE 13
fitness for the purpose, seaworthiness of the Unit or of its value for sale, insurance 13.1. - These General Conditions constitute the sole contractual obligations binding
or chartering. together the Society and the Client, to the exclusion of all other representation,
5.3. - The Society does not declare the acceptance or commissioning of a Unit, nor statements, terms, conditions whether express or implied. They may be varied in
of its construction in conformity with its design, that being the exclusive writing by mutual agreement.
responsibility of its owner or builder, respectively. 13.2. - The invalidity of one or more stipulations of the present General Conditions does not
affect the validity of the remaining provisions.
13.3. - The definitions herein take precedence over any definitions serving the same
purpose which may appear in other documents issued by the Society.

BV Mod. Ad. ME 545 j – 16 February 2004

Contents

1 Introduction 4
1.1 Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Starspec structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Theoretical Background 8
2.1 Spectral Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Short Term Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Second order spectral analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Wave Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4.1 JONSWAP Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.4.2 Wallop Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4.3 Gamma spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4.4 Ochi-Hubble spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.4.5 Gauss spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5 Directional Spreading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5.1 Formulation "cos n" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5.2 Formulation "cos2s" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5.3 Formulation "Wrapped normal" . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5.4 Formulation equivalences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.6 Quadratic roll damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.7 Long Term Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.7.1 Counting Cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.7.2 Long term probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.8 Fatigue Damage Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.8.1 Short term damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.8.2 Long term damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.9 Correction to the fatigue damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.10 Intermittent wetting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.11 Operational criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.12 Equivalent Design Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.12.1 Regular Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.12.2 Response Conditioned Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.12.3 Directional Response Conditioned Wave . . . . . . . . . . . . . . . . . . . . . . . . 23

1 / 67 Bureau Veritas Research Department

 CONTENTS

3 Input file 24

3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.2 RAOs files and path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.2.1 Paths and files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.2.2 Note : Water depth and Speed override . . . . . . . . . . . . . . . . . . . . . . . . 28

3.2.3 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.3 Sea-states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.3.1 Scatter diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3.3.2 Directional scatter diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3.3.3 List of Sea-states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3.3.4 List of Sea-states, new format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

3.3.5 Spectrum type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

3.3.6 Tabulated spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.3.7 Azimuth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3.4 S-N curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.5 Advanced parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3.5.1 Operational criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3.5.2 Reference duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5.3 RAO interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5.4 Integration steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.6 Quadratic roll damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3.7 Intermittent wetting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3.8 Design waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3.9 Compression correction for damage calculation . . . . . . . . . . . . . . . . . . . . . . . . 40

3.10 Output option . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

3.11 Combining several project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

3.12 Time domain output (experimental feature) . . . . . . . . . . . . . . . . . . . . . . . . . . 44

2 / 67 Bureau Veritas Research Department

 CONTENTS

4 Computation and output files 45

4.1 Short term results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

4.2 Long term results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

4.3 Fatigue results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

4.4 Group results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

4.5 EDW results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

4.6 Operational results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

4.7 Wave spectrum results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

4.8 RAOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

4.9 Spreadsheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

4.10 Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

4.11 Log file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

4.12 Gnuplot from Hydrostar/Starspec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

5 Example 55

5.1 Damage calculation with scatter diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

5.2 Round trip in a bimodal sea-states list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

A List of keywords 61

B HSlps 63

C Operation on RAOs 65

D RAO format 66

3 / 67 Bureau Veritas Research Department

Chapter 1

Introduction

Starspec is a spectral analysis software that performs short term and long term response analysis. Addi-
tionally, given a S − N curve for a structural detail, it performs the fatigue damage calculation by using
Miner’s Rule.

Starspec calculates:

• Spectral Moments;
• Short term response;
• Long term response;
• Fatigue damage through a given S × N curve;

The program usage is described in chapter 3 and 4. For in depth description about how Starspec works,
theoretical details are provided in chapter 2.

Some test examples can be found in the folder “/examples” and provide the user the possibility of using
and modifying the input files for tutorial purposes. To run the test examples the user may follow the
instructions provided in the following chapters of this manual.

If there’s any problem in installing or using Starspec please contact the following support people at Bureau
Veritas Research Department:

e-mail: [email protected]

1.1 Installation

The Starspec_setup performs what is necessary to run the program.

1.2 Starspec structure

Starspec is structured with one module, which computes the different results depending on the data
provided in the input file.

4 / 67 Bureau Veritas Research Department

 CHAPTER 1. INTRODUCTION

1.3 Conventions

The following conventions should be used:

• Sea states are described by their significant height (HS ), peak period (TP ) in s and direction (β)
in degree in geographical reference (i.e. from north, 0◦ is north, 90◦ is east )

Waves direction

0° - North

270° 90°
West East

180° - South

Figure 1.1: Waves direction βgeo

• The azimuth of the vessel is specified from north (clockwise)

5 / 67 Bureau Veritas Research Department

 CHAPTER 1. INTRODUCTION

North

Azimuth

Figure 1.2: Azimuth α

• RAOs are defined with respect to the circular wave frequency (not encounter frequency). The RAOs
are defined in HydroStar reference which is represented on Figure 1.3. (180◦ is head-seas, 90◦ is
starboard)

6 / 67 Bureau Veritas Research Department

 CHAPTER 1. INTRODUCTION

270°
315° 225°

x
0
0° 180°

ap M fp

45° 135°
90°

Figure 1.3: Relative incidence for RAOs βRAO

Note : The relation ship between the wave direction βgeo the azimuth α and the relative incidence in
the vessels reference βHydrostar is :

βRAO = −βgeo + α + 180 (1.1)

7 / 67 Bureau Veritas Research Department

Chapter 2

Theoretical Background

The aim of Starspec v1.0 is to perform the first order spectral analysis, short term and long term statistics and
fatigue damage calculations. This chapter presents the theoretical background of the software. More detail
can be found in text book, or published papers [2].

2.1 Spectral Analysis

We consider the spectral density Sw (ω) representing the distribution in frequency (ω) of the wave energy and
the RAO(ω) as being the transfer function of any first order quantity, like motions, accelerations, relative
wave elevation, stress, etc.

The spectral density of response can be then defined as:

SR (ω) = RAO2 (ω) ∗ Sw (ω) (2.1)

The spectral moments can be defined as:

Z ∞
mn = ω n SR (ω)dω (2.2)
0

If several spectra with different directions are used, the spectral momentum are sum :

m Z
X ∞
mn = ω n Sw (ω, θi ) ∗ RAO2 (ω, θi )dω (2.3)
i ω=0

or more generally :

Z Z ∞
mn = ω n Sw (ω, θ) ∗ RAO2 (ω, θ)dωdθ (2.4)
θ ω=0

8 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

Some statistical parameters can then be obtained. The mean period Tm , often referred as T1 , is defined as:

m0
Tm = 2π (2.5)
m1

TZ , which corresponds to the mean zero up-crossing period is given by:

r
m0
TZ = 2π = T2 (2.6)
m2

The mean frequency of maxima is

r
1 m4
µ= (2.7)
2π m2

The bandwidth of the response spectrum can be characterize by :

m22
2 = 1 − (2.8)
m0 m4

=0 w Narrow banded spectrum

=1 w Wide banded spectrum

Warning : For several spectra like JONSWAP, the moment of order 4 does not converge, so that  = 1.
The results given by Starspec come from a numerical integration, the m4 value, and the variable related to
m4 are thus not accurate.

Effect of forward speed

For the calculation of the spectral moments with forward speed, the frequency to be taken into account shall
not be the wave frequency but the encounter frequency defined as:

ωe = ω − kV cosβ (2.9)

ωe the encounter frequency,

ω the wave frequency,
V the forward speed in m/s
βRAO the wave incidence
k the wave number

In finite water depth the dispersion equation has to be solved, whereas in infinite water-depth, the wave number
is given by :

ω2
k= (2.10)
g

It can be easily demonstrated that the area below the wave spectrum does not change with the forward speed.
Thus,the wave m0 remains the same. However, the other moments change.
NOTE: The RAO is always given in wave circular frequency.

9 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.2 Short Term Statistics

The short term corresponds to the duration of one sea state (typically 3 hours), which is considered to be
stationary.

We consider a random variable R being the range of response. Assuming that the process is narrow banded,
the probability density of response range follows the Rayleigh’s distribution:

−R2
 
R
p(R) = exp (2.11)
4m0 8m0

The distribution function being :

−R2
 
P (R) = 1 − exp (2.12)
8m0

Significant response

From equation 2.11, the mean value of the (1/n) highest cycles can be calculated. For instance :

√
R1/2 = 2.50 m0 (2.13a)
√
R1/3 = 4.00 m0 (2.13b)
√
R1/10 = 5.10 m0 (2.13c)

The value R1/3 is often called significant value RS .

Maximum definitions

The definition of short term maxima can often be confusing, as many different definition are used in the
industry.

• Maximal Short term response exceeded with a risk α over a sea-state duration Dss

Over N cycles the repartition function is given by

−R2
P (R) = (1 − e 8m0 )N (2.14)

with, α = 1 − P (R) , it comes (given in double amplitude):

q
Rmax (Dss , RTz , α) = 2 · 2m0 ln( (1−α)−1
1/N −1 ) (2.15)

N being given by N = Dss /RTz

example : with α = 0.5, Dss = 10800s and RTz = 10s ==> Rmax = 1.90 ∗ Rs

10 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

• Short term mean maximum

The average of the maximum occurring in a sea-state is, for large value of N, given by (in double
amplitude) :
√ √
R̃max (Dss , RTz ) = 2· ( 2 ln N + √ γ ) m0 (2.16)
2 ln N

N being given by N = Dss /RTz

Where γ is the Euler constant ( γ = 0.5772 ). The standard deviation of the maximum is :
π √
σRmax = √ m0 (2.17)
3 ln N

example : with Dss = 10800s and RTz = 10s , ==> R̃max = 1.94 ∗ Rs
• Response (here in double amplitude) exceeded with a risk β over one cycle.
On one cycle, the probability of exceedance is directly given by eq 2.12 :
q
Rprob β = 8m0 ln β1 (2.18)

1
example : With β = 1000 ==> Rprob 1/1000 = 1.86 ∗ Rs
Note that over a duration Dss corresponding to N cycles, taking β = 1/N gives Rprob 1/N =
√
8m0 ln N . This value is exceeded with a risk α = 0.63 !

2.3 Second order spectral analysis

The second order spectrum is defined by :

Z ∞
SR2 (∆ω) = 8 ∗ Sw (ω) ∗ Sw (ω + ∆ω) ∗ |QT F (ω, ω + ∆ω)|2 dω (2.19)
0

The second-order moments are then given by :

Z ∞
mn = ω n SR2 (ω)dω (2.20)
0

The mean value being calculated by :

Z ∞
R̃2 = 2 ∗ Sw (ω) ∗ QT F (ω, ω)dω (2.21)
0

2.4 Wave Spectra

Four types of wave spectra may be used: JONSWAP, Wallops, Ochi-Hubble and Gauss. Custom (tabulated)
spectra can also be input.

11 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.4.1 JONSWAP Spectrum

The JONSWAP formulation is based on an extensive wave measurement program known as the JOint North
Sea WAve Project. The spectrum represents wind-generated seas with fetch limitation. The formulation is
more general and includes the spectrum of Pierson Moskowitz as a particular case. It can be written as:

"  −4 #   −(ω−ωP )2 

5 2 4 −5 5 ω exp
2σ 2 ω 2
Sw (ω) = A ∗ Hs ωp ω exp − γ P (2.22)
16 4 ωP

γ is the peak-enhancement factor, the effect of which is to increase the peak of Pierson- Moskovitz
spectrum. (γ = 1 corresponds to a Pierson- Moskovitz spectrum)
σ is a relative measure of the width of the peak.

In most cases, σ = 0.07 for ω < ωP and σ = 0.09 for ω > ωP

In Starspec, the parameter α is fitted to the peak enhancement factor γ through the relationship (ISO-19901-
1):

1
A= (2.23)
5· (0.065· γ 0.803 + 0.135)

The error is less than 1.5%, up to gamma=30. (0.1% at gamma=10)

Regression for period conversion (valid up to γ = 15) :

Tz √
= 0.6063 + 0.1164 ∗ γ − 0.01224 ∗ γ (2.24)
Tp

Tm √
= 0.6687 + 0.1182 ∗ γ − 0.01489 ∗ γ (2.25)
Tp

T0m1 √
= 0.7844 + 0.08316 ∗ γ − 0.01033 ∗ γ (2.26)
Tp

Exact value for γ = 1 (Pierson-Moskovitz spectrum)

Tz 5π
= ( )−1/4 ≈ 0.71 ≈ 1/1.408 (2.27)
Tp 4

√
Tm 2 ∗ 5−1/4
= ≈ 0.772 ≈ 1/1.296 (2.28)
Tp Γ(3/4)

√ 3 1


T0m1 2 · 54 Γ 4
= ≈ 0.857 ≈ 1/1.1665 (2.29)
Tp 20

12 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.4.2 Wallop Spectrum

The Wallops wave spectrum is a generalized Pierson-Moskowitz spectrum. It has 2 shape-parameters, m and
q, instead of one for Pierson-Moskowitz:

−m " −q #

H2
 
ω ω
Sω (ω) = A S exp −B (2.30)
ωP ωP ωP

  m−1
q
 
m q m m−1
with B = q and A = 16 q /Γ q .

The wallops spectrum is equivalent to the Pierson-Moskovitz spectrum for m = 5 and q = 4.

2.4.3 Gamma spectrum

Gamma spectrum is a generalized spectrum, that embeds Wallop, Pierson-Moskowvitz and Jonswap or
Torsethaugen spectrum as particular cases [3].

N −1
  
N −(ω−ωP )2
M(M ) M Hs 2 ω − N ( ω )−M exp
∗ ( )−N ∗ e M ωp
2 2
Sw (ω) = Aγ N −1
∗ ∗γ 2σ ω
P (2.31)
Γ( M ) 16ωp ωp

Aγ is the normalizing factor. In practice, the following regression is used :

0.1
4.1 ∗ (N − 2M 0.28 + 5.3)(0.96−1.45M )
(ln γ)f2 + 1
Aγ = (2.32)
γ

with
0.37
f2 = (2.2M −3.3 + 0.57)N (0.53−0.58M )
+ 0.94 − 1.04M −1.9 (2.33)

Particular cases are :

• Jonswap : M = 4 , N = 5
• Wallop : γ = 1 , M = q , N = m
• Ochihubble : N = 4λ + 1 , M = 4 , γ = 1
• Pierson-Moskowitz (ITTC / Breitshneider) : N = 5 , M = 4 , γ = 1
• Torsethaugen : N , M and γ values given in [3].

Note : Like Ochi-Hubble spectrum, Torsethaugen spectrum usually has two components (wind-sea and swell).
In StarSpec, considering the two components as two spectrum of the same sea-state allows to take into account
for the directionality of the component separately.

13 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.4.4 Ochi-Hubble spectrum

The Ochi-Hubble spectrum has a "bimodal" shape, and is actually made of two Wallop spectrum (presented
above).
The energy density function of the Ochi-Hubble spectrum is defined as follows:

2 λi 2
HS,i
     
1X 1 4 1  ωp,i 4
Sω (ω) = λi + ωp,i exp − λi + (2.34)
4 i=1 4 Γ(λi )ω 4λi +1 4 ω

where:
ω is the circular frequency (rad/s)
HS is the significant wave height
ωp is the peak wave circular frequency
λ, is the spectral shape parameter.

The equivalence between Ochi-Hubble and Wallops is given for q = 4 and mi = 4λi + 1.

2.4.5 Gauss spectrum

The Gaussian spectrum is defined by its significant wave height, its peak period Tp and its standard deviation
σ

Hs2 1 (ω−ωp )2
Sω (ω) = · √ e− 2σ2 (2.35)
16 σ 2π

The following relationships can be derived :


H2


m0 = 16s

m1 = ωp



m2 = σ 2 + ωp2 (2.36)
m4 = ωp4 + 6σ 2 ωp2 + 3σ 4





 q
σ = ωz2 − ωp2


2.5 Directional Spreading

To represent the angular distribution of the wave, "directional" spectra Sw (ω, β) is used :

S(ω, β) = S(ω)G(ω, β) (2.37)

where the spreading function G(ω, β) satisfies:

Z 2π
G(ω, β)dβ = 1 (2.38)
0

In Starspec, the spreading is not considered frequency dependent, so that G(ω, β) = G(β):

14 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.5.1 Formulation "cos n"

π
G1 (β) = C1 (n)cosn (β − β̄) |β − β̄| ≤ (2.39a)
2
π
G1 (β) = 0 |β − β̄| ≥ (2.39b)
2

β̄ being the mean propagation direction

Rπ
The coefficient C1 (s) is a normalizing factor determined by the relation −π
G(θ)dθ = 1. It is given by :

Γ(n/2 + 1)
C1 (n) = √ (2.40)
πΓ(n/2 + 1/2)

NOTE:The function cosn is the spreading function in IACS Rec.34, for which n = 2.

2.5.2 Formulation "cos2s"

β − β̄
G2 (β) = C2 (s)cos2s − π ≤ β − β̄ ≤ π (2.41)
2

β̄ being the mean propagation direction

Rπ
The coefficient C2 (s) is a normalizing factor determined by the relation −π
G(θ)dθ = 1. It is given by :

Γ(s + 1)
C2 (s) = √ (2.42)
2 πΓ(s + 1/2)

2.5.3 Formulation "Wrapped normal"

+∞
1 X (β̄ − β − 2kπ)2
Gwm (β) = √ exp(− 2
) (2.43)
σwm 2π −∞
2σwm

15 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.5.4 Formulation equivalences

In practice, the shapes of the different formulations are very similar with the following relationship between
coefficients :

1 4
σ2 ' ' with σ in radians (2.44)
n+1 2s + 1

Spreading formulation
Mean angle = 180.0 deg
0.9
COSN N=4
COS2S S=8
0.8 Wrapped Normal SIGMA = 25.6 deg

0.7

0.6

0.5

0.4

0.3

0.2

0.1

-0.1
0 50 100 150 200 250 300 350 400
Angle (deg)

Figure 2.1: Spreading formulation

2.6 Quadratic roll damping

The roll damping is known to be non-linear and is often modelled as quadratic. To keep the calculation in the
frequency domain, this damping has to be linearised.
For irregular waves, stochastic linearization is used :

r
8
Beq = Blin + σ Bquad (2.45)
π θ̇

Where σθ̇ is the standard deviation of the roll velocity.

Z ∞
σθ̇2 = ω 2 ∗ Roll(ω)2 ∗ SR (ω) dω = m2 (2.46)
2

In practice, several RAOs are provided for different linear damping. The equation 2.45 is then solved numeri-
cally, the convergence is reached after a few iterations.
In Starspec, the regula-falsi scheme is used to solve Beq (m2 (Beq )) − Beq = 0

16 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.7 Long Term Statistics

Considering that a short term analysis, as above described, is performed for a list of sea states observed during
a reference period DREF . The long term distribution can then be obtained by cumulating the results from
the short term analysis. There are different methods to perform the long term statistics calculation. In this
section, the method implemented in Starspec is described.

2.7.1 Counting Cycles

The method consists in counting, over all sea-states, of all maxima of the response (i. e. each response
cycle). This method, identical to the Battjes wave counting, is commonly used for the evaluation of structural
response of ships and offshore structures under cyclic (wave) actions, for both fatigue and extremes.It can be
written as:

SS=N
XSS
nex (X) = nSS (1 − P (X)) (2.47)
SS=1

where,

NSS is the total number of sea states ,

nex (X) is the expected number of exceedance of a response level X, over a reference duration DREF ,
P (X) is the Rayleigh distribution for the sea-state SS according to 2.12,
nSS is number of response cycles for a sea state SS, defined as (DREF ×365.24×24×3600)/RTZ ×
prob(SS); prob(SS) being the probability of occurrence of the sea-state.

2.7.2 Long term probability

If the long-term is long enough, all the waves can be considered independent, and the long term cumulative
distribution can then be written as :

N ss
Y 2
P (R) = (1 − e−R /8m0 nss
) (2.48)
ss

This equation can then be solved to get the range exceeded with a risk α in a given duration (α = 1 − P (R)).
(This is the long term equivalent to the short term maximum given by the equation 2.15 )

17 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.8 Fatigue Damage Calculation

The fatigue calculation in Starspec in performed through a Miner’s Rules. A S − N curve is defined by the
user according to the applicable standards and described by the parameters m and K. For a given response
range S, the maximum number of cycles until the rupture is given by :

K
N (S) = (2.49)
Sm

Several slopes can be defined, the relationship is then :

Ki
N (S) = f or SQi−1 < S < SQi (2.50)
S mi

18 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

log S

K 1 , m1

SQ 1

K 2 ,m2

SQ 2
K 3 , m3
S Q Threshold 
n

NQ 1
NQ 2
log N

19 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.8.1 Short term damage

The damage DAMsht induced by one sea-state (duration Dsht , response up-crossing period RTz ) can be
written as :

∞
X n(S)
DAMsht = (2.51)
N (S)
S=0

n(S) is the number of cycle with the range S

where :
N (S) is the number of cycle to failure for a given Stress range S

Using 2.49 and 2.51 combine with a Rayleigh probability density of the maximum (2.11) the short term damage
is given by :

∞
S m+1 − 8m
Z
Dsht S2
DAMsht = · e 0 dS (2.52)
RTz · K 0 4m0

R∞
which can be written, using the incomplete gamma function : Γ(a, x) = x
ua−1 e−u du

√
Dsht 2 m
DAMsht = ·( Rs )m Γ(1 + , 0) (2.53)
RTz · K 2 2

In case of multi slopes S − N curve, as described on the figure 2.2, the short term damage may be written as
:

Nslope √ 2 2
Dsht X 1 2 mi 2 SQ mi 2 SQi−1
DAMsht = ( Rs )mi [Γ(1 + , i
) − Γ(1 + , )] (2.54)
RTz i=1 Ki 2 2 Rs2 2 Rs2

With :

(
SQ0 =∞
(2.55)
SQN slope = T hreshold

2.8.2 Long term damage

The long term damage DAMlt is the sum of the short term damage associated with a probability :

Dref ∗ 3600 ∗ 24 ∗ 365.24 X

DAMlt = DAMsht · prob(SS) (2.56)
Dsht

Note : Eventually, the long term damage does not depend of sea-states duration Dsht

20 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.9 Correction to the fatigue damage

Empirical studies has shown that fatigue damage was reduced when part of the cycle was in compression. This
effect is taken into account in classification societies rules (example ). Usually a factor 0.6 is applied to the
compression part of the cycle. Several assumptions can be made to assess the amount of stress cycle which is
in compression, two are implemented in StarSpec.

• Reduced damage due to compression for welded elements (depends on limit of elasticity Re)
For those welded elements, residual stresses equal to yield stress and no relaxation are assumed. So the
stress cycle maxima is the limit of elasticity.

Figure 2.3: Stress cycle for welded elements

(
∆S 0 = ∆S ∗ Kc
(2.57)
Kc = 0.4Re
∆S + 0.6 , bounded in [0.8,1.0]

• Reduced damage due to compression for non-welded elements (depends on the mean stress)
For such elements, it is assumed that stress cycles around its mean (still water) value.

21 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

Figure 2.4: Stress cycle for non-welded elements

Noting ρ the reduction factor due to compression (usually taken as 0.6), M the mean stress, the corrected
range R̄ is then given by :

1+ρ
R̄ = R + M · (1 − ρ) (2.58)
2

Note : Both correction relies on a modification of the long term stress distribution. Thus, analytical integration
(2.54) of damage per sea-state can not be used. Instead, a numerical integration of the modified long term
distribution with the sn-curve is used (in the output file, the user should thus look at the columns with the
labels "NUMERICAL").

2.10 Intermittent wetting

Due to linear approach of classic spectral calculation, the calculated damage to elements near the free-surface
is not satisfactory. For instance, an element just above the mean free-surface would never be subjected to
wave pressure.
To address this issue, the method StarSpec use a method based on equivalent RAOs, with a linearization in
irregular waves.

2.11 Operational criteria

In Starspec, operational criteria can be defined. When one of these criteria is exceeded, the speed or heading
of the ship are changed in order to reduce the response of the ship regarding those criteria.
The scheme used in Starspec is quite basic :
-If several speed data are available, the speed of the ship is reduced.
-If allow by the user, the azimuth of the vessel is modified.
-If the criteria are not met, the speed and azimuth are those minimizing the response (with regards to the
criteria).

22 / 67 Bureau Veritas Research Department

 CHAPTER 2. THEORETICAL BACKGROUND

2.12 Equivalent Design Wave

2.12.1 Regular Wave

2.12.2 Response Conditioned Wave

Note: On a given sea-state, none of the response can exceed its value at the probability level used for the
EDW.

(
X
Ai = Sw(ωi ) ∗ |RAOL (ωi , β)| m dωi
0 (2.59)
φi = −RAOφ (ωi , β)

√
Loads value at p X = SHTl = coef ∗ m0L

Value of any responses on this design wave :

X
rEDWs = Ai |RAOs (ω, β)| cos(φs − φl )
X
EDWs <= Ai |RAOs (ω, β)|
X X
EDWs <= Sw(ωi ) ∗ |RAOL (ωi , β)|
dωi |RAOs (ω, β)|
m0
X coef
EDWs <= Sw(ωi ) ∗ |RAOL (ωi , β)||RAOs (ω, β)| √ dωi
m0L
P
Sw(ωi ) ∗ |RAOL (ωi , β)||RAOs (ω, β)|coef dωi (2.60)
EDWs <= pP
Sw(ω)|RAOL (ωi , β)|2 dωi
pP pP
Sw(ωi ) ∗ |RAOL (ωi , β)|2 dωi Sw(ωi ) ∗ |RAOS (ωi , β)|2 dωi coef
EDWs <= pP
Sw(ω)|RAOL (ωi , β)|2 dωi
qX
EDWs <= Sw(ωi ) ∗ |RAOS (ωi , β)|2 dωi coef
EDWs <= SHTs

2.12.3 Directional Response Conditioned Wave

23 / 67 Bureau Veritas Research Department

Chapter 3

Input file

Depending on the data provided in the input file, the program computes different kind of results. The following
data are relevant to the spectral calculation :

• RAOs data (Must be specified)

• Sea-states description (Must be specified)

• Sea-states probability (If long term computation is required)
• S − N curve (If fatigue calculation is required)

The user have also some option, which have default values :

• The range and step of integration;

• The reference duration for the long term computation ;
• The Sea-state duration for the short term statistics;
• The output desired

The comment character is #. Every input file has to be finished by the keyword :

ENDFILE

24 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

25 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.1 Overview

RAO_PATH Path to the RAOs

C:\Dvt\\modules\hspec\input\rao
ENDRAO_PATH

RAO_FILE RAO file

My1.RAO SAF 4
My2.RAO SAF 10
My3.RAO SAF 5 Amplification factor
ENDRAO_FILE

SPECTRE_TYPE Spectrum description

JONSWAP
SPREADING COS2S 8 COS2S 2
ENDSPECTRE_TYPE
List of sea-states, number of modes
LIST_SEASTATE NBMODE 2
#Dir 1 Hs1 Tp1 g1 Dir2 Hs2 Tp2 g2 Comments line (#)
313.81 0.52 9.39 1.33 353.66 1.65 6.34 1.21
312.59 0.52 9.40 1.33 358.37 1.60 5.83 1.15
336.35 0.51 9.39 1.38 21.42 1.71 5.75 0.84
332.48 0.48 9.38 1.40 18.19 1.88 6.27 1.07
ENDLIST_SEASTATE

AZIMUTH Azimuth of the vessel and probability

0
1

DREF 100 Reference duration, in years

SN_CURVE SN-curve
0.1 5 4.330E15
53.4 3 1.520E12
ENDSN_CURVE

OUTPUT OUTPUT Section

LONGTERM DETAIL PNG
SHORTTERM DETAIL EXCEL
PLOTX
RAO PNG
RPERIOD 20
EXCEL
ENDOUTPUT

ENDFILE

26 / 67 Bureau Veritas Research Department

Figure 3.1: 3D response graph
 CHAPTER 3. INPUT FILE

3.2 RAOs files and path

3.2.1 Paths and files

Path :

First, the paths where the RAOs to be analyzed are stored have to be specified between the keywords :
RAO_PATH and ENDRAO_PATH. In general, one path is provided, but if the computation involves more
than one speed, each folder contains the RAOs for one speed. The files in each folders have to have the same
names. The folders have to be listed from the slowest speed to the highest.

To associate a predefined probability to each path, the required keyword is PATH_PROB, followed by the
list of probability. By default, the probability is set to 1 for the highest speed, and 0 for the other ones.

Note 1: If operational criteria are used, the probability of each speed will be modified independently for each
sea-states.

Note 2: The program was designed to use different path for different speeds, it is however possible to use
different path for different loading conditions.

Note 3: By default the path is set to /rao

Files :

Then the files you want to use should be listed between the keywords : RAO_FILE and ENDRAO_FILE.

For each RAO, a amplification factor can be applied with the keyword SAF, the name of the RAO is defined
after the keyword NAME. By default, the name of the RAO is the name of the file without the extension.
The name of the RAOs is used to labeled the output and also to defined advanced computation parameters.
If computed on 0-180◦ , the RAO can be symmetrized with the keyword SYM 1.

The following options can be applied to the RAO (keywords should be added next to the RAO file name):

Keyword Description Default Stable/beta

SAF saf Multiply the RAO by a coefficient 1
SYM id Symmetrize the RAO 0
WATERDEPTH depth Defined the Water-depth RAO file value
NAME name label of the RAO rao file name
SN_CURVE id Associated SN_CURVE 0
FILTER_WE we maximum encounter frequency (after which RAO=0) disabled
TRUNC_WE we maximum encounter frequency (after which RAO=cnst) disabled
SCALE scale Scale the RAO disabled beta
CRIT maxRs Operational criteria disabled
EDW Design wave (see 3.8) disabled beta
EDWH β Design wave heading Auto beta
LCF Load combinaison factor (see 3.8) disabled beta
BLIN blin roll linear damping disabled
BQUAD bquad roll quadratic damping disabled
MEAN_CORRECTION mean stress correction (see 2.9) disabled
BVRULES_B743 Re compression correction for welded elements (see 2.9) disabled
INTERWET ctrl ID Intermittent wetting (2.10) disabled
RWE ID RWE for intermittent wetting disabled
GROUP ID Attach the transfer function to a group (see 4.4) disabled

27 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

QTF input :

If second order calculation is desired, the QTF file names have to be specified between the keywords :
QTF_FILE and ENDQTF_FILE.

3.2.2 Note : Water depth and Speed override

For each RAOs, related informations like speed or waterdepth are directly read in the RAOs file.

However, if not defined in the RAO file, the user is able to set them within the Starspec input file with the
keyword PATH_SPEED and WATERDEPTH.

If different speeds are defined in the input file and in the RAO file, a warning is displayed and the calculation
stops.

If different water depths are defined in the input file and in the RAO file, a warning is displayed and the
calculation goes on, using the water depth used in the input file (in order to allow the user to bypass shallow
water option, which could sometimes be time consuming).

The speed is given in m/s and the water depth in m.

WATERDEPTH keyword should follow to the RAO file name.

As the speed is related to the path and not to the rao file, the keyword PATH_SPEED should be next to
RAO_PATH.

Once again, this is necessary only if the RAO file does not contain the information.

If speed or water depth are not specified in the RAO file neither in the input file, the default values are :

• Speed = 0.0 m/s

• Water depth = Inf.

Note : The use of WATERDEPTH is thus rather limited... Water depth information is needed only if :
-the vessel has a forward speed,
-the water depth is not infinite
-the water depth is not mentioned in the RAO file.

28 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.2.3 Example

Eventually the part of the input file where the RAOs are specified should look like :

RAO_PATH PATH_SPEED 2.5

c:\starspec exemple25
ENDRAO_PATH

RAO_FILE
pitch012B.rao NAME Pitch_ballast
roll.rao NAME Roll_ballast
heave.rao NAME Heave_ballast
surge.rao NAME Surge_ballast
FY.rao SAF 50.0 NAME Stress_ballast
ENDRAO_FILE

If different speeds with associated probability are provided, the input could be something like :

RAO_PATH PATH_SPEED 2.5 5.2 9.4 PATH_PROB 0.33 0.33 0.33

c:\starspec exemple25
c:\starspec exemple52
c:\starspec exemple94
ENDRAO_PATH

RAO_FILE
pitch.rao
roll.rao
heave.rao
surge.rao
FY.rao SAF 50.0
ENDRAO_FILE

Note : The keyword PATH_SPEED is used in the example but is generally not required, the speed being
read in the RAO files of the corresponding folder.

3.3 Sea-states

Sea-states should be given in the geographic reference. (See section 1.3)

Sea-states data can be provided in several way.

• List of Sea-states, possibility of multi-modal calculation

• Scatter diagram, identical in every direction
• Directional scatter diagram

29 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.3.1 Scatter diagram

The scatter diagram is defined between the keywords SCATTER and ENDSCATTER. SCATTER has to be
followed by the size of the diagram NB_HS number of Hs and NB_TP Number of Tp.

The scatter diagram should then look like this :

SCATTER NB_HS 4 NB_TP 3

5 7 9
1.0 5 10 4
2.0 7 12 8
3.0 8 13 6
4.0 0 5 3
ENDSCATTER

NOTE : For JONSWAP spectrum, it is also possible to define scatter diagram in Tz instead of Tp. This is
done by changing NB_TP by NB_TZ. Tp is then computed from Tz with the following regression formula,
valid for 0 < γ < 10

Tz
Tp = (3.1)
0.6063 + 0.1164γ 0.5 − 0.01224γ

The wave direction is defined with the keyword WAVEDIR_PROB :

WAVEDIR_PROB
Direction 1 Direction 2 [...] Direction n
Probability 1 Probability 2 [...] Probability n

If all wave directions have the same probability, it is possible to specify only the number of directions desired.
The syntax is then :

WAVEDIR_PROB ISO number of wave directions

3.3.2 Directional scatter diagram

The keywords are SCATTER and ENDSCATTER. But, in addition to NB_HS number of Hs and NB_TP
Number of Tp, the number of wave directions has to be specified next to NB_DIR. Then, the wave direction
of each scatter diagram have to be specified.

The directional scatter diagram should then looks like this :

SCATTER NB_HS 4 NB_TP 3 NB_DIR 2

180.
5 7 9
1.0 5 10 4
2.0 7 12 8
3.0 8 13 6
4.0 0 5 3

30 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

210.
5 7 9
1.0 5 8 4
2.0 6 12 8
3.0 8 13 6
4.0 0 5 3
ENDSCATTER

3.3.3 List of Sea-states

A list of sea-states can be defined between the keywords LIST_SEASTATE and ENDLIST_SEASTATE.

If the sea-states are not unidirectional, the number of mode should be given after the keyword NBMODE

The sea-states are then described. The last parameters is the probability of the sea-states. If no probability is
given, it is assumed that the sea-states have the same probability.

LIST_SEASTATE NBMODE 1
Wave direction Hs Tp Spectrum_paramters_1 [Spectrum_paramters_2...] prob
Wave direction Hs Tp Spectrum_paramters_1 [Spectrum_paramters_2...] prob
Wave direction Hs Tp Spectrum_paramters_1 [Spectrum_paramters_2...] prob
...
ENDLIST_SEASTATE

If the spectrum parameters are already given in the section SPECTRE_TYPE, they should be skipped in
the list.

If several direction are defined, the spectrum description is repeated "nbmode" times.

Note : The probability of each sea-states is automatically made adimensional (total = 1).

The keyword DELTATP_SECOND or DELTATP_PERCENT, next to LIST_SEASTATE allow the user

to defined an interval of Tp (T p − deltatp ; T p + deltatp). The number of value desired is defined next to
the keyword VALUE.

Example :

LIST_SEASTATE DELTATP_SECOND 1 VALUE 3

150 2.5 7.2 3.3
ENDLIST_SEASTATE

is equivalent to

LIST_SEASTATE
150 2.5 7.2 3.3
150 2.5 8.2 3.3
150 2.5 6.2 3.3
ENDLIST_SEASTATE

31 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.3.4 List of Sea-states, new format

List of sea-state, as specified in 3.3.3 has some limitations : the spectrum type, the number of modes and
spreading is the same for all sea-state. To overcome this limitations, a new format is introduced, starting
from HydroStar v8.1. List of sea-states can now be specified in between the keywords LIST_GENERIC
and ENDLIST_GENERIC. Each line corresponds to the full description of the sea-states (meaning that
the section "SPECTRUM_TYPE" has then no use). For each spectrum, all relevant parameters should be
specified using the relevant keywords, as defined in table 3.1. For each spectrum wave heading is specified
with "WHEADING", and spreading can be added (COSN, COS2S or WNORMAL keywords). The number
of spectrum on each sea-state (line) is not limited.

JONSWAP HS TP GAMMA
GAMMASPEC HS TP GAMMA M N
SIMPLEOCHIHUBBLE HS TP LAMBDA
GAUSS HS TP SIGMA
WALLOP HS TP M Q
SPECTRUMFILE filename

Table 3.1: Available spectra and associated parameter

Example

LIST_GENERIC
JONSWAP HS 1.5 TP 12.0 WHEADING 150. COSN 2
JONSWAP HS 1.5 TP 12.0 GAMMA 1.0 WHEADING 150. JONSWAP HS 1.5 TP 12.0 GAMMA 2.0 WHEADING 120.
ENDLIST_GENERIC

3.3.5 Spectrum type

The description of the spectrum to be used is defined between the keywords SPECTRE_TYPE and END-
SPECTRE_TYPE. The type of spectrum desired is set using one of the keywords : JONSWAP, WALLOP,
GAUSS. To change the default value of σa = 0.07 and σb = 0.09 in the Jonswap spectrum, the keyword
SIGJONSWAP can be used.

If scatter diagram are used, this spectrum type has to be followed by thes parameters :

JONSWAP γ
SIGJONSWAP γ σa σb
WALLOP m q
S_OCHI_HUBBLE λ
OCHI_HUBBLE λ1 Hs2 Tp2 λ2
GAUSS σ
GAMMASPEC γ M N
WHITE_NOISE ω1 ω2 ω3 ω4

If a list of sea-states is defined, the parameters of the spectrum can be defined either globally for all spectrum
or individually. If set globally, the spectrum parameters have to be defined as above and repeated for each
mode. If set individually, the parameters have then to be defined in the list of sea-states.

32 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

If different kind of spectrum are used, all the parameters have to be set for each sea-states.

The spreading has to be specified after the keyword SPREADING, the formulation is then chosen with the
keyword COS2S, COSN or WNORMAL, followed by s, n or σ respectively (σ in degree).

Note : When using spreading, the angular integration step has to be carefully chosen (See NB_STEP)

Example :

• Scatter Diagram

SPECTRE_TYPE
JONSWAP 3.3
SPREADING COS2S 4
ENDSPECTRE_TYPE

SCATTER
...
ENDSCATTER

• List of bimodal sea-states, spectrum parameters set individually

SPECTRE_TYPE
JONSWAP
SPREADING COS2S 4 COS2S 8
ENDSPECTRE_TYPE

LIST_SEASTATES NBMODE 2
180 5.4 15.0 3.3 150 5.4 10.0 1.0 0.5
172 4.4 14.0 3.3 150 5.4 10.0 1.0 0.5
ENDLIST_SEASTATES

• Same list of bimodal sea-states, but parameters set globally

SPECTRE_TYPE
JONSWAP 3.3 1.0
SPREADING COS2S 4 COS2S 8
ENDSPECTRE_TYPE

LIST_SEASTATES NBMODE 2
180 5.4 15.0 150 5.4 10.0 0.5
172 4.4 14.0 150 5.4 10.0 0.5
ENDLIST_SEASTATES

33 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

• Bimodal sea-states with different kinds of spectrum

SPECTRE_TYPE
JONSWAP GAUSS
SPREADING COS2S 4 COS2S 8
ENDSPECTRE_TYPE

LIST_SEASTATES NBMODE 2
180 5.4 15.0 3.3 150 5.4 10.0 0.10 0.5
172 4.4 14.0 3.3 150 5.4 10.0 0.08 0.5
ENDLIST_SEASTATES

3.3.6 Tabulated spectra

In top of parametric spectrum models, it is possible it input arbitrary, tabulated, 2D spectra. The syntax is
the following :

LIST_SEASTATE SPECTRUM3D NBMODE 24 METEO_PATH RAO/meteo

ss_0.txt
...
ss_n.txt
ENDLIST_SEASTATE

where

NBMODE is the directional discretization (number of heading in the spectrum file)

METEO_PATH is the path to the tabulated spectra files

Each tabulated spectra is stored in indiviudal text files, with the following format :

w ω1 ... ωn
β1 Sw (ω1 , β1 ) ... Sw (ωn , β1 )
...
βn Sw (ω1 , βn ) ... Sw (ωn , βn )

2
m
where Sw (ω, β) is the spectral density ( rad·rad/s ) at circular frequency ω (rad/s) and heading β (degree).

34 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.3.7 Azimuth

Once the sea-states have been defined, the azimuth of the vessel is required. (Detail in Section 1.3 p. 5 )

The azimuth of the vessel is set with the keyword AZIMUTH; the first line is the list of azimuth, the second
the list of associated probability.

For instance :

LIST_SEASTATES NB_MODE 2
180. 1. 14.08 1.0 210. 1. 14.08 3.3 1.0
150. 1. 14.08 1.0 180. 1. 14.08 3.3 1.0
ENDLIST_SEASTATES

AZIMUTH
30
1

is equivalent to

LIST_SEASTATES NB_MODE 2
150. 1. 14.08 1.0 180. 1. 14.08 3.3 1.0
120. 1. 14.08 1.0 150. 1. 14.08 3.3 1.0
ENDLIST_SEASTATES

AZIMUTH
0
1

It is also possible to defined several azimuth with associated probability :

Example :

AZIMUTH
30 60 90
0.3 0.3 0.4

In this example, the azimuth 30◦ has a 0.3 probability and will be referred as "Azimuth 1" in the output file.

If all the azimuths have the same probability, it is possible to specify the number of azimuth desired. The
syntax is then :

AZIMUTH ISO number of azimuth

35 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

Note : Defining both wave direction and azimuth is flexible but somewhere redundant, the essential data
being the incidence of the wave in the vessel reference.

Thus, for undetermined wave incidence, the following input file would provide the same results :

WAVEDIR_PROB ISO 72
AZIMUTH
0

or

WAVEDIR_PROB
0
1
AZIMUTH ISO 72

An other way to get the same result, which have to be AVOIDED (72 times more computation), would be to
distributed equiprobably both wave direction and azimuth.

Concerning the output, Starspec would provide more convenient files with the second solution. Indeed, as
the azimuth can be used with both scatter diagram and list of sea-states, this information prevails the wave
direction.

3.4 S-N curve

The S −N curve can be given with several slopes. The different slopes parameters should be provided between
the keywords : SN_CURVE and ENDSN_CURVE

SN_CURVE
S(i), m(i), K(i) range of response (threshold of curve S − N ), m and K
..
.
S(nb_c),m(nb_c),K(nb_c)
ENDSN_CURVE

The S − N curve thus should look likes this :

SN_CURVE
53.4 3 1.520e12
0.01 5 4.330e15
ENDSN_CURVE

As the S − N curve data are redundant, the program check the consistence of the data and warn if they are
not.

It is also possible to define several S − N curves associated with different RAOs. The S − N curve use by
defaut is the one defined with the keyword SN_CURVE alone. The specific S − N curve are defined with
the same keyword followed by the number of the S − N curve (must be numbered from 1).

36 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

The Association RAOs / S − N curve is defined in the RAO section.

Example :
RAO_FILE
elem1.rao SN_CURVE 1
elem2.rao
ENDRAO_FILE

SN_CURVE
53.4 3 1.520e12
0.01 5 4.330e15
ENDSN_CURVE

SN_CURVE 1
73.62 3 3.99e12
0.1 5 2.162e16
ENDSN_CURVE

The RAOs "elem1" will be computed with the S − N curve 1 and "elem2" with the default S − N curve.

3.5 Advanced parameters

The keywords in this section are not compulsory.

3.5.1 Operational criteria

If different speeds are provided, operational criteria can be specified for each RAO. If no user probability is
defined, the speed is set to the maximum for every sea-state and is reduced if the criteria are not met.
The criteria defined in StarSpec is relative to the Significant Response Rs
The value of the criteria has to be defined with the keyword CRIT next to the RAO name.
If the vessel is allowed to change its azimuth, the keyword is
ALLOW_AZIM_CHANGE NSTEP Number of step in 360◦
The option ALLOW_AZIM_CHANGE is not compatible with PATH_PROB.
Example :
RAO_FILE
Roll.rao CRIT 8
Pitch.rao CRIT 6
ENDRAO_FILE

ALLOW_AZIM_CHANGE NSTEP 72

It is also possible to use the Ochi criteria for slamming events. The syntax is then :
RAO_FILE
Zrelative_motion.rao SLAMCRIT draugh prob length
ENDRAO_FILE

37 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.5.2 Reference duration

The default reference duration DREF is 20 years, it can be changed :

DREF reference duration

The reference duration define :

• The duration in which the fatigue damage is provided

• The duration in which the long term distribution is plotted

• The duration associated to the probability "DREF_PROBMAX"

DREF does not change the long term value defined in term of cycle probability or return period (PROB and
RPERIOD keywords).

3.5.3 RAO interpolation

Linear and spline interpolation are availabe in Starspec. This is set with the following keywords :

F_INTERPOL for frequency interpolation

H_INTERPOL for heading interpolation

followed by the type of interpolation wanted :

LINEAR for linear interpolation.

NATURAL for natural cubic spline interpolation.
OVERHAUSER for overhauser (Catmul-Rom) cubic spline interpolation.
HERMITE for "Hermite" cubic spline interpolation.
The Hermite spline implemented here use an evaluation of the derivatives such as the resulting spline does
not overshoot the data point.

By default, the interpolation is linear.

3.5.4 Integration steps

The following keywords can be used to set different integration parameters:

OMEGA ωmin ωmax Integration range Default : 0.1 and 6..0

STEP ∆ω Integration step Default : 0.0025
NB_HSTEP NB_step Number of direction steps Default : 72

38 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.6 Quadratic roll damping

Quadratic roll damping is used as soon as the keyword BQUAD is find next to a RAOs parameters. Then
ROLL_COUPLING have to be specified if you want all the RAOs to depend of the roll damping.

If quadratic roll damping is used, the roll RAO is identified in the RAO list with the keywords defining the
damping coefficients. This RAO must contain data with different linear damping values. The value of the
linear damping and quadratic damping are defined respectively after the keywords, BLIN and BQUAD.

If ROLL_COUPLING option is enabled, all the RAO files must contain data with different linear damping.

If a RAO file contains several damping and if the quadratic roll damping option is disabled, the first damping
listed in the RAO file is used.

Example :

RAO_FILE
Roll.rao BLIN 0.7836e + 9 BQUAD 0.114e + 12
ENDRAO_FILE

ROLL_COUPLING

It is possible to use quadratic roll damping and operational criteria at the same time.

Note : The Roll RAO is assumed to be given in degree (which is general practice). However, the damping
coefficient are given in [N m/(rad/s)] for the linear part (BLIN) and [N m/(rad/s)2 ] for the quadratic part
(BQUAD).

3.7 Intermittent wetting

The elements subjected to intermittent wetting must be identified with the keyword INTERWET (see 3.2.1),
then the ID of the RWE rao (which controls the linearisation) has to be specified. The element RAOs subjected
to intermittent wetting should contains different blocks of data, corresponding to different footprint height.

The control RAOs (Relative Wave Elevation RAO) are identified thanks to the keyword RWE, followed by the
chosen ID.

3.8 Design waves

Starspec can calculates design wave for a selection of RAOs and associated Load Combinaison Factor (LCF)
for another set of RAOs. The RAOs for which design waves have to be computed are identified thanks to the
keyword EDW. The RAOs for which LCFs have to be calculated (on all the EDW) are identified thanks to
the keyword LCF.

Starspec systematically computes 3 types of design waves :

• Regular waves

(
X
A= |RAO(ω,β)|
(3.2)
φ = −RAOφ (ω, β)

39 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

• Response Conditioned Waves

(
X
Ai = Sw(ωi ) ∗ |RAO(ωi , β)| m dωi
0 (3.3)
φi = −RAOφ (ωi , β)

• Directional Response Conditioned Waves

(
X
Aij = Sw(ωi , βj ) ∗ |RAO(ωi , βj )| m dωi dβj
0 (3.4)
φij = −RAOφ (ωi , βj )

For each of these design waves, the parameters (heading, frequency, wave spectrum) can be done in many
ways, leading to more or less accurate results [1]. The way to select these parameters can be chosen with the
keyword EDWOPTION :

Regular RCW DRCW

Heading | Frequency Heading | Sw Sw
1 MaxRAO | MaxRAO MaxRAO | SS_maxContrib SS_maxContrib
2 WeightedSpec | MaxRAO WeightedSpec | SS_maxContrib SS_maxContrib
3 WeightedSpec | MaxRAO WeightedSpec | WeightedSpec(ihead) WeightedSpec
4 WeightedSpec | WeightedSpec WeightedSpec | WeightedSpec WeightedSpec

where :
MaxRAO stands for the location where the RAO is maximum.
WeightedSpec is the cumulated 3D response spectrum weighted by the contribution to the long term value.
SS_maxContrib is the spectrum shape of the sea-state contributing the most to the long term value

All the results about design waves will be found in the subfolder "EDW" (see section ??).

Note 1 : When using option 1, the range in which to look for the maximum can be defined by the keyword
"EDWREG_WRANGE wmin wmax".

Note 2 : Option 1 and 2 are available for wave data input as scatter diagram only.

Note 3 : In order to optimize design wave for further time domain calculation, option 2 use the non-
interpolated RAOs for the irregular wave calculation.

3.9 Compression correction for damage calculation

To elements RAO on which fatigue correction has to be applied are identified thanks to the keyword presented
in 3.2.1.

To enable the correction, the output option FATIGUE NUMERICAL has to be specified (the correction can
not be applied with the analytical calculation of the damage).

40 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

3.10 Output option

Only three file are always generated : report.csv, report.dat and the log file. The other output are controlled
by the user.

The output options have to be placed between the KEYWORDS OUTPUT, ENDOUTPUT

RPERIOD Return period Specific return period (related to 2.47), see 3.2
PROB Probability Specific probability (related to 2.47, 3.2)
WAVESPECTRUM FORMAT PLOT3D Wave spectrum
RAO FORMAT PLOT3D INT RAOs (input data)
LONGTERM FORMAT DETAIL EXCEL long term distribution
SHORTTERM FORMAT DETAIL EXCEL Short term results
DSS_PROBMAX probmax Short term maximum probability
DREF_PROBMAX probmax Long term maximum probability (related to 2.48)
FATIGUE YES/NO DETAIL Fatigue results
OPERATION FORMAT HSLIM Operability polar
PLOTX Results plotted along the ship
COEFF coefficient Output of Rs * Coeff
EQWAVE Output of the equivalent design wave
EQSS Output of the "equivalent sea-state"
EXCEL ’All.csv’ file

Figure 3.2: Long term distribution and extreme values

FORMAT can take the value :

41 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

blank gnuplot file

PNG png file
PS postscript file
JPG jpg file
NO no output

For FATIGUE LONGTERM and the additional option DETAIL is available only if a scatter diagram is input,
this option result in a scatter diagram for longterm and fatigue.

The option INT for the RAOs ask the output of the interpolated RAOs.

The option EXCEL and DETAIL output detail of the short term response in text format and .csv respectively.
excel.

The keyword DSS_PROBMAX allows the output of short term maxima, related to the sea-states duration
(eq 2.15).The value is in the excel files in the folder Short_term.

The option HSLIM (for operability polar plot) displays the limiting Hs with regard to the criteria. (The default
polar is the ratio between the significant response and the criteria.)

The results of the options EQSS, EQWAVE and COEFF are output through additional information in the
files ’report.csv’ and ’report.dat’

Several probability or return period can be specified, detailed contribution are however output only for the first
one.

The default output are :

• Wave spectrum : No
• RAOs : No
• Long term : long term distribution and Gnuplot file only

• Short term : Results in report.csv only

• Fatigue : Total damage only
• Report and ’csv ’ file

NOTE 1 : The output option RPERIOD and PROB are exclusive. Be aware that linking a probability to a
return period without calculation implies an assumption on the response period.

NOTE 2 : Some picture exports require gnuplot 4.2 or higher and are not supported by gnuplot 4.0.

42 / 67 Bureau Veritas Research Department

 CHAPTER 3. INPUT FILE

Example :

3Dlt.gnu

Lt_raoname.png

OUTPUT

Lt_distribution.csv
WAVESPECTRUM PNG

RPERIOD 20

rao_name_case.png

LONGTERM DETAIL PNG EXCEL

SHORTTERM PNG EXCEL

Project_moments.csv

PLOTX

sht_plotx.gnu lt_plotx.gnu
RAO PNG

FATIGUE DETAIL

3Ddam.gnu
EXCEL

ENDOUTPUT
All.csv

Figure 3.3: Output section

43 / 67 Bureau Veritas Research Department
 CHAPTER 3. INPUT FILE

3.11 Combining several project

If long term calculation have been performed for several projects, it is possible to combine them with associated
probability.

The RAOs to be used have still to be defined. The projects to be combined are listed between the keywords
COMBINE and ENDCOMBINE. The probability of each project is defined next to the project name.

The input file should thus look like :

RAO_FILE
myRAO1.rao
myRAO2.rao
ENDRAO_FILE

COMBINE
projet1 0.2
projet2 0.2
projet3 0.6
ENDCOMBINE

ENDFILE

NOTE : If the keyword COMBINE is in the input file, no other calculation are performed.

3.12 Time domain output (experimental feature)

StarSpec allows the user to output some time domain reconstruction.

TIMEDOMAIN [WRITEWIF WRITEDIST WRITETIME] Enable time domain reconstruction

WRITEWIF Output spectral response with phases
WRITEDIST Output distribution of maxima
WRITETIME Output time series of response
TIME_NBSEED Set the number of seed

The duration of the simulation is set by DSS. The results are stored in "Time_domain" subfolder.

44 / 67 Bureau Veritas Research Department

Chapter 4

Computation and output files

Starspec can be launch from the command windows by "Hspec input.spc myProject"

Once the computation has been performed, results are stored in the folder /myProject_spec The log file
myProject.log contains a list of all the warning which had appeared during the computation as well as an echo
of the input file.

Depending on the calculation, one or several of these sub-folders are created :

• Short_term
• Long_term

• RAOs
• Fatigue
• Wavespectrum
• Operation

Three files are always created :

• report.dat
• report.csv
• myProject_spec.log

45 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

4.1 Short term results

The short term results are stored in the file projname_moments.dat located in the folder Short_term. The
EXCEL option write into a format directly read by excel (projname_moments.csv). The values output are
the following (depending on the option, some columns might not appear):
Label Description
raofile Name of the RAO
x x coordinates
raopath path of the RAO
NSS id of the seastate
Azimuth Azimuth of the ship
m0 Moment of order 0
m1 Moment of order 1
m2 Moment of order 2
m4 Moment of order 4 (take with caution)
Rs Siginificant response (double amplitude)
Rtz Response up-crossing period
AmpMeanMaxSht Mean maximum short term extrem ( = R̃max /2 eq 2.16)
AmpMaxSht Maximum short term assicated with a risk and a duration (= Rmax /2 eq 2.15)
epsilon Spectrum bandwitdth ( eq 2.8, take with caution)
mu Spectrum bandwitdth ( eq 2.7 ,take with caution)
Wave direction Wave direction of the sea-state
Hs significant height of the sea-state
Tp Peak period of the sea-state
Gamma γ value (for Jonswap spectrum only)
Spreading cosn spreading parameter
The spectral moment (m0, m1, m2, m4), the significant response and some other statistics are provided.

If the sea-states are input as a scatter diagram, a 3D representation of the significant response can be generated.

Figure 4.1: 3D response graph

46 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

4.2 Long term results

The long term results for all RAOs are stored in the files lt_distribution.dat located in the folder Long_term. A
shorter file containing only the data at a specified probability or return period is lt_rperiod.dat or lt_prob.dat

A gnuplot file (lt.gnu) is also written.

Long term distribution

1000
82ps1 Azimuth 000
82ps1 Azimuth 030
900 82ps1 Azimuth 060
82ps1 Azimuth 090
82ps1 Azimuth 120
800 Value = 0.8118E+03 82ps1 Azimuth 150
82ps1 Azimuth 180
82ps1 Azimuth 210
700 82ps1 Azimuth 240
RANGE (double amplitude)

82ps1 Azimuth 270

82ps1 Azimuth 300
600 82ps1 Azimuth 330
82ps1 rperiod 0.2500E+02
82ps1 total
500

400

300

200

100

0
0.1 1 10 100 1000 10000 100000 1e+006 1e+007 1e+008 1e+009
Nb of cycles in 25.00 years

Figure 4.2: Long term distribution

If the sea-states are input as a scatter diagram, and if a specific probability or return period has been provided,
the contribution of the different Hs and Tp to the long term value is given(First summed over all the azimuth
then for each azimuth).

Azimuth contribution to the long term value

0.8
Nb of cycles in 25.0 years 82ps1

0.7

0.6

0.5
Nb of cycles

0.4

0.3

0.2

0.1

0
0 30 60 90 120 150 180 210 240 270 300 330 360
Azimuth

Figure 4.3: Contribution to the long term value

47 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

Contribution to the extreme value

Azimuth 90. Wave dir : 0. euronavfr82ps1.rao

0.06
0.05
0.04
0.03
0.02
0.06 18
0.05 0.01
16
0.04 0
Nb of cycles 14
0.03
0.02 12
0.01 10
0 8 Hs
6
4
0 5 2
10 15 20 25 0
30
Tp

Figure 4.4: Contribution to the long term value

4.3 Fatigue results

The Fatigue folder contains the results of the damage calculation. Depending, of the input sea-states, the file
myProject_damage_detail.dat would be slightly different. If scatter diagrams have been specified, the file
contains the scatter digram of damage, if a list of sea-states have been set, the file contains the list of the
damage.

If the sea-states are input as a scatter diagram, and if a specific probability or return period has been provided,
the contribution of the different Hs and Tp to the damage is given; first summed over all the azimuth then
for each azimuth.

48 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

Significant response

euronavfr83ps1.rao 330.

0.18
0.16
0.18 0.14
0.16 0.12
0.14 0.1
0.12 0.08
0.1 0.06
Damage
0.08 0.04
0.06 0.02
0.04 0
0.02
0
0
2
4
6
8
10 Hs
12
14
30 25 20 16
15 10 5 0 18
Tp

Figure 4.5: 3D fatigue scatter diagram

Note : The damage written is those files is defined according to the reference duration.

4.4 Group results

If RAOs have been associated to a group (see 3.2.1), a additional file named "group_projectName.csv" is
created. The file contains, for each group, the maximum longterm value and damage over all RAOs of the
group.

4.5 EDW results

When EDW option is used, folder "EDW" is created. This folder contains

• a CSV file storing all LCFs on all EDWs

• a "wif" file for each EDW. The "wif" file define a unique wave time series by its discrete spectrum that
includes phases
• time trace for all EDWs and LCFs

49 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

4.6 Operational results

If operational criteria are defined, the following results can be output :

• Speed polar (speed in polar coordinates)

Figure 4.6: Polar

• "Operability scatter diagram" integrated through the azimuth.

50 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

Operationability
18 3

16
2.5
14

12 2

10
Hs (m)

1.5
8

6 1

4
0.5
2

0 0
0 5 10 15 20 25 30
Tp (s)

Figure 4.7: Operational scatter diagram

0 - The criteria are met.

1 - The criteria are met if the speed is reduced.
2 - The criteria are met is the vessel azimuth is changed.
3 - It is not possible to meet the criteria.

• "Operational scatter diagram" detailed for each azimuth.

51 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

Azimuth : 120.0000 Wavedir : 0.0000

Speed
18 14

16 12

14
10
12
8
10
Hs (m)

6
8
4
6
2
4

2 0

0 -2
0 5 10 15 20 25 30
Tp (s)

Figure 4.8: Operational scatter diagram

A negative value corresponds to a change of the vessel azimuth.

4.7 Wave spectrum results

The Wavespectrum folder contains the wave spectrum used for the calculation, a gnuplot file is also available

52 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

Wave Spectrum
0.4
Tp = 2.11
Tp = 3.52
Tp = 4.93
0.35 Tp = 6.34
Tp = 7.74
Tp = 9.15
Tp = 10.56
0.3 Tp = 11.97
Tp = 13.38
Tp = 14.78
Tp = 16.19
0.25
每⼀條線都是不規則波 Tp = 17.60
Tp = 19.01
Tp越⼩，頻率峰值越⼤ Tp = 20.42
S(w)

0.2 Tp = 21.82
Tp = 23.23
Tp = 24.64
Tp = 26.05
0.15 Tp = 26.05

0.1

0.05

0
1 2 3 4 5 6
Wave frequency (rad/s)

Figure 4.9: Wave spectra

4.8 RAOs

The folders RAOs contains the RAOs which have been read and re-write by Starspec as well as the one
interpolated and extrapolated. Two gnuplot files are created for both set of files.

4.9 Spreadsheet

The "csv" files All.csv contains the following results :

Description Label
Significant response Rs
Response up-crossing period Rtz
Long term value Long term value
Fatigue life Fatigue life
Number of cycles exceeding the long term value in each sea-state NB cycles
Damage during each sea-state Damage
Whether or not the ship is "operational" Operational

By default, all the results for all the RAOs are in the file. To limit the size of the table, it is possible to
restrict the "csv" files to a few RAOs. The RAOs names to be output should then be between the keywords
RAO_OUT and ENDRAO_OUT. This option is also applied the long term and short term ’csv’ files.

53 / 67 Bureau Veritas Research Department

 CHAPTER 4. COMPUTATION AND OUTPUT FILES

4.10 Report

A summary of all the results available in the file "report.dat".

This summary contains (if computed) :

• Echo of the input file.

• The mean speed of the ship
• Maximum short term response for each RAOs (and the corresponding sea-state)

• Long term value for each RAOs.

• Fatigue life for each RAOs.

The file ’report.csv’ contains the same results, but easier to use with a spreadsheet.

4.11 Log file

The log file contains the warning message which have been printed during the calculation.

4.12 Gnuplot from Hydrostar/Starspec

All the graph presented in the previous section are in the folders created by Starspec. It is also possible to
plot the graphs directly from Starspec thanks to the options :

• -sht for short term results.

• -lt for the long term distribution.

• -3Dlt for the detail of the long term value.
• -dam for the detail of the damage.

• -pol for the speed polar.

• -rao for the rao.
• -ws for the wave spectrum.
• -xsht for the short term value along the ship

• -xlt for the long term value along the ship

So to get the long term distribution, the command to type from Hydrostar is "Starspec -lt". If Starspec is run
outside Hydrostar, this would be "Starspec -lt projet".

Note : All the results to be plotted have of course to be computed first... Those commands are just shortcuts
to Gnuplot

54 / 67 Bureau Veritas Research Department

Chapter 5

Example

5.1 Damage calculation with scatter diagram

In this example, the aim is to compute a fatigue computation in sea-states described by a scatter-diagram.
The stress RAOs is given for 5 elements of the ship, the ship is considered with 6.095 m/s speed, regardless
of the sea-states.

The input file is the following :

55 / 67 Bureau Veritas Research Department

 CHAPTER 5. EXAMPLE

RAO_PATH PATH_SPEED 6.095

C:\Dvt\Hydrostar sur Drsvn\trunk\modules\hspec\input\rao\RAO_full_load
ENDRAO_PATH

RAO_FILE
euronav_fr82ps1.rao
euronav_fr82ps2.rao
euronav_fr82sb1.rao
euronav_fr82sb2.rao
euronav_fr83ps1.rao
ENDRAO_FILE

SPECTRE_TYPE
JONSWAP 1.0
ENDSPECTRE_TYPE

SCATTER NB_HS 17 NB_TP 18

2.112 3.52 4.928 6.336 [...]
0.5 0.0 0.0 1.3 133.7 [...]
1.5 0.0 0.0 0.0 29.3 [...]
2.5 0.0 0.0 0.0 2.2 [...]
[...] [...] [...] [...] [...]
ENDSCATTER

WAVEDIR_PROB
0
1.0

AZIMUTH ISO 12

SN_CURVE
53.4 3 1.520E12
0.01 5 4.330E15
ENDSN_CURVE

DREF 25.0
NB_HSTEP 72
D_SEASTATE 10800

OUTPUT
FATIGUE DETAIL
WAVESPECTRUM PNG
SHORTTERM DETAIL EPS
LONGTERM DETAIL EPS
RAO PNG
PROB 1e-8
ENDOUTPUT

ENDFILE

56 / 67 Bureau Veritas Research Department

 CHAPTER 5. EXAMPLE

Note : The scatter diagram in the input file is truncated for clarity

The results are :

Long term distribution

1000
82ps1 Azimuth 000
82ps1 Azimuth 030
900 82ps1 Azimuth 060
82ps1 Azimuth 090
Value = 0.8238E+03 82ps1 Azimuth 120
800 82ps1 Azimuth 150
82ps1 Azimuth 180
82ps1 Azimuth 210
700 82ps1 Azimuth 240
RANGE (double amplitude)

82ps1 Azimuth 270

82ps1 Azimuth 300
600 82ps1 Azimuth 330
82ps1 prob 0.1000E-07
82ps1 total
500

400

300

200

100

0
0.1 1 10 100 1000 10000 100000 1e+006 1e+007 1e+008 1e+009
Nb of cycles in 25.00 years

Figure 5.2: Long term results

Significant response

euronavfr83ps1.rao 330.

0.18
0.16
0.18 0.14
0.16 0.12
0.14 0.1
0.12 0.08
0.1 0.06
Damage
0.08 0.04
0.06 0.02
0.04 0
0.02
0
0
2
4
6
8
10 Hs
12
14
30 25 20 16
15 10 5 0 18
Tp

Figure 5.3: 3D fatigue graph

57 / 67 Bureau Veritas Research Department

 CHAPTER 5. EXAMPLE

5.2 Round trip in a bimodal sea-states list

In this example, the aim is to compute the long term value of relative wave elevation (here for a return period
of 15 years). The ship performs round trip in a bimodal sea-states.

The input file is the following :

58 / 67 Bureau Veritas Research Department

 CHAPTER 5. EXAMPLE

RAO_PATH
C:\Dvt\Hydrostar sur Drsvn\trunk\modules\hspec\input\rao\RAO_full_load
ENDRAO_PATH

RAO_FILE HYDROSTAR
rwe5.rao
ENDRAO_FILE

SPECTRE_TYPE
JONSWAP
SPREADING COSN 8 COSN 8
ENDSPECTRE_TYPE

LIST_SEASTATE NBMODE 2
167.0 1.51 7.10 3.00 198.00 0.88 4.25 1.50
158.7 1.25 6.83 3.00 189.70 0.87 4.26 1.50
156.5 1.07 6.63 3.00 185.50 0.92 4.55 1.50
160.4 0.98 6.61 3.00 188.40 0.97 4.60 1.50
167.8 1.08 6.41 3.00 192.80 0.82 4.16 1.50
164.1 1.12 6.29 3.00 187.10 0.68 3.77 1.50
170.3 1.00 6.80 3.00 189.30 0.78 4.21 1.50
160.5 0.82 7.15 3.00 172.50 0.95 5.16 1.50
157.8 0.78 7.15 3.00 174.80 1.04 5.14 1.50
83.5 0.70 12.88 4.00 181.50 1.22 5.10 1.50
107.4 0.95 12.97 4.00 179.40 1.04 4.62 1.50
347.0 1.00 12.97 4.00 154.00 1.06 4.69 1.50
249.0 1.16 12.88 4.00 143.00 0.87 4.29 1.50
246.7 1.18 12.71 3.00 166.70 1.02 4.68 1.50
218.1 1.26 12.47 3.00 164.10 1.04 5.03 1.50
199.7 1.39 11.96 3.00 176.70 1.26 5.48 1.50
175.3 1.94 11.36 3.00 171.30 1.20 5.20 1.50
174.3 2.37 10.59 3.00 183.30 1.25 5.54 1.50
153.4 2.73 10.55 3.00 173.40 1.04 5.13 1.50
ENDLIST_SEASTATE

AZIMUTH
45 225
0.5 0.5

NB_HSTEP 72
D_SEASTATE 10800
NB_PROC 8
DREF 100.

OUTPUT
RAO PNG
LONGTERM EPS
RPERIOD 15
ENDOUTPUT

ENDFILE

59 / 67 Bureau Veritas Research Department

 CHAPTER 5. EXAMPLE

On the graph plotted on figure 5.5, the contribution of the 2 different ways are plotted, as well as the total of
the round trip.

Long term distribution

7
rwe5 Azimuth 045
rwe5 Azimuth 225
rwe5 rperiod 0.1500E+02
Extrem Value = 6.123 rwe5 total
6

5
RANGE (double amplitude)

0
0.1 1 10 100 1000 10000 100000 1e+006 1e+007 1e+008 1e+009
Nb of cycles in 100.00 years

Figure 5.5: long term distribution

60 / 67 Bureau Veritas Research Department

Appendix A

List of keywords

RAO

RAO_PATH
ENDRAO_PATH
RAO_FILE
ENDRAO_FILE
PATH_SPEED
PATH_PROB
WATERDEPTH
SAF

Sea-states

NBMODE
SPECTRE_TYPE
WAVEDIR_PROB
LIST_SEA
SCATTER
NB_HS
NB_TP
NB_TZ
NB_DIR
AZIMUTH
ISO

Option

STEP
DREF
OMEG
NB_HSTEP
DSS
ALLOW_AZIM_CHANGE
ROLL_COUPLING
CRIT
BLIN
BQUAD

61 / 67 Bureau Veritas Research Department

 APPENDIX A. LIST OF KEYWORDS

Fatigue
SN_CURVE

Output

RAO_OUT ENDRAO_OUT OUTPUT

ENDOUTPUT
WAVESPECTRUM
JPG
PS
PNG
RAO
INT
EXCEL
DETAIL
PROB
RPERIOD
LONGTERM
SHORTTERM
OPERATION
COEFF
PRINT

Additional keywords

GNUPLOT_PATH : Path to Gnuplot

LTRNB : Discretization of the long term distribution
DEBUG : Help debugging the input files, is not affected by ’#’
NB_PROC : Set the number of processor
FROM_HEADING_ANALYSIS : The azimuth is given for each sea-state
REL_INC : The data are then given in the local Hydrostar reference
D_SEASTATE : Set the duration of sea-states, only relevant for short term results
SLAMCRIT : Ochi criteria for slamming probability

62 / 67 Bureau Veritas Research Department

Appendix B

HSlps

HSlps is a little script made to generate and launch HydroStar input files. It is convenient for quadratic roll
damping purpose as it can define a loop on the damping coefficient (an RAO file containing all the data with
the linear damping is also generated).

Figure B.1: HSlps input file example

The input file has to be in the same folder as the HydroStar "original" input files. The hslps script can be
called from command line with hslps input.lps.
The keyword NB_PROC is used to defined how many processor are wanted for multi-threaded HydroStar
module (for instance HSrdf)
All commands are optional (i.e. if the radiation/diffraction has already been performed, the mechanical
(HSMCN) calculations and RAOs output can be run without having to rerun hsrdf). If hsrdf has been launch
outside HSlps, the keyword RDF_DONE should be added. The diffraction/radiation files of the current
folder is then used.

Figure B.2: HSlps scheme

63 / 67 Bureau Veritas Research Department

 APPENDIX B. HSLPS

HSloop options

LAUNCH YES/NO : yes –> launch the computation / no –> create the input files
NO_SUB : No subfolder is created (does not work if a speed loop is defined
CLEAN 1/2/3 : Remove temporary files
DAMPING_LOOP / END_DAMPING_LOOP : Define a damping loop, apply to "LINVISCOUSDAMPING"
MCN_KEY (next to DAMPING_LOOP) : Define a loop on a "MCN_KEY" keyword that would
be located in HSmcn input.
SPEED_LOOP / END_SPEED_LOOP : Define a speed loop
NB_PROC nbproc : Number of processors to be used

Hydrostar commands available

HSLEC
HSRDF
HSTNK
HSMCN
HSPRS
HSWLD
HSQTF
HSRAO

64 / 67 Bureau Veritas Research Department

Appendix C

Operation on RAOs

HSope is a module to handle HydroStar RAOs (Addition, Multiplication...). The syntax is the following :

RAO_IN <raoin_path>
file1.rao
file2.rao
ENDRAO_IN

OPERATION
file4.rao = file2.rao OPERATOR2 file1.rao
file5.rao = file4.rao OPERATOR1
ENDOPERATION

RAO_OUT <raoout_path>
file4.rao
file5.rao
ENDRAO_OUT

The following operators are available :

RAO_ADD : Add another RAO

RAO_MINUS : Subtract another RAO
SCAL_MULT : Multiply by a scalar
RAO_MULT : Multiply by another RAO
DERIVATE : Derivate (Multiply by the encounter circular frequency * i)
INTEGRATE : Integrate (Divide by the encounter circular frequency * i)
PS_PSSB : RAO ps/sb 0-180 –> RAO ps 0-360
SB_PSSB : RAO ps/sb 0-180 –> RAO sb 0-360
SYM : Symmetrize(/anti-symmetrize) the RAO
SCALE : Scale the RAO
FILTER_WE : Put 0 where ωe > omega
TRUNC_WE : Put the last value if ωe > omega
BANDPASS_WE : Put 0 where ωe < ωl or ωe > ωu (ex rao2 = rao1 BANDPASS_WE 0.5 1.0)
LINEAR : LINEAR combination (ex rao = LINEAR 0.5 rao1 0.5 rao2)

Note that all operations are, of course, done using complex arithmetic (phase are accounted for).

65 / 67 Bureau Veritas Research Department

Appendix D

RAO format

The RAO format read by StarSpec is HydroStar format. The RAO file is constituted with a header and the
RAO data. The header contains the following data :

Water depth # Waterdepth : 50.0000

Forward speed # Forward speed : 0.0000 m/s
Coordinates # COORD 159.307 0.001 -5.143
RAO type # RAOTYPE "GROLL
Complex type # AMP/PHASE
Number of headings # NBHEADING 13
Heading list # HEADING 0.00 15.00 30.00 ...

The data are then written in the following format

ω1 RAO(ω1 , heading1 )...RAO(ω1 , headingn )P HI(ω1 , heading1 )...P HI(ω1 , headingn )

...
...
ωn

66 / 67 Bureau Veritas Research Department

Bibliography

[1] Guillaume de Hauteclocque, Quentin Derbanne, and Amine El-Gharbaoui. Comparison of different equiva-
lent design waves with spectral analysis. In ASME 2012 31st International Conference on Ocean, Offshore
and Arctic Engineering (OMAE), number 83405, pages 353–361, 2012.
[2] B. Molin. Hydrodynamique des structures offshore. Editions Technip, 2002.

[3] Knut Torsethaugen, Sverre Haver, et al. Simplified double peak spectral model for ocean waves. In The
Fourteenth International Offshore and Polar Engineering Conference. International Society of Offshore and
Polar Engineers, 2004.

67 / 67 Bureau Veritas Research Department