|
|
Technical PapersThe following are technical papers written by Formation Design Systems' staff members and presented at various conferences and seminars worldwide. Please feel free to download copies of the pdf files.
Optimisation of Vessel Resistance using Genetic AlgorithmsAndrew Mason, Patrick Couser, Garth Mason, Cameron R. Smith, Brian R. von KonskyCOMPIT'05, Hamburg, Germany, May 2005. AbstractIt has been found that an artificial neural network is able to produce results of sufficient accuracy to be useful for preliminary prediction of vessel resistance, with the major benefits of: being relatively simple to set up; being easily retrained with new data; and that Froude number may be easily included as an independent variable. In this work the Neural Network is fitted directly to the original tank test data, rather than to a set of smoothed curves varying in the Froude number axis. In addition, different network architectures have been investigated, with networks containing two hidden layers being seen to perform well. A procedure for the optimisation of design parameters using a Genetic Algorithm has also been evaluated. The Genetic Algorithm uses the approximation provided by the Neural Network response surfaces for its objective function. This has been found to be effective and of acceptable performance.
@inproceedings{MasonEtAl2005, Artificial Neural Networks for Hull Resistance PredictionAndrew Mason, Patrick Couser, Garth Mason, Cameron R. Smith, Brian R. von KonskyCOMPIT'04, Siguenza, Spain, April 2004. AbstractThe applicability of artificial neural networks to the problem of ship resistance prediction as an alternative to more traditional statistical regression models has been investigated. In this work, an artificial neural network has been used as an interpolation tool to predict the residuary resistance of a systematic series of catamaran forms. It has been found that an artificial neural network is able to produce results of sufficient accuracy to be useful for preliminary prediction of vessel resistance, with the major benefits of: being relatively simple to set up; being easily retrained with new data; and that Froude number may be easily included as an independent variable.
@inproceedings{CouserEtAl2004, On the Effect of Tank Free Surfaces on Vessel Static StabilityPatrick CouserInternational Journal of Maritime Engineering, Part A3, Volume 146, 2004. AbstractThis paper examines the effect of fluid free surfaces in slack tanks on a vessel’s transverse, static stability. The exact, transverse movement of fluid in a half-full, rectangular tank and its effect on vessel stability is derived. This is compared with results from computer software that also models the static position of the fluid in the tank, including the effects of both heel and trim; the traditional correction to vertical centre of gravity based on upright second moment of area of the tank waterplane; and the IMO free surface moment method as described in IMO A.749(18) and MSC.75(69)[1, 2, §3.3]. Once validated, the application of the computer software has been demonstrated by using it to examine the free surface effect of tanks at filling conditions other than 50%.
@article{Couser2004, A Software Developer's Perspective Of Stability CriteriaPatrick CouserSTAB 2003, 8th International Conference on the Stability of Ships and Ocean Vehicles, Madrid, Spain, September 2003. AbstractThe assessment of vessel stability to specified stability criteria is now required for virtually all forms of water-craft. This is routinely done using computer software that models the vessel’s hydrostatic properties and evaluates its stability against specified criteria. Over the last 24 months, Formation Design Systems has implemented a wide rage of stability criteria in its hydrostatics analysis computer program Hydromax.This paper describes some of the outcomes, observations and questions arising from this software development cycle, from a software development and naval architecture perspective. It provides a viewpoint of stability criteria which is different from that of both regulatory bodies (such as IMO) and naval architects dealing with the day-to-day aspects of vessel design. It also presents some cross-fertilisation ideas from the software industry that could enhance the understanding of stability criteria. It is hoped that the suggestions made in this paper can lead to clearer, more concise and more accurate stability criteria which will, in turn, lead to safer vessels. Many stability criteria and the way in which vessels’ hydrostatic properties are evaluated and presented are inherited from the pre-desktop computer era. Now that desktop computers are commonplace in the design office, evaluation of hydrostatic properties and stability criteria can be made more accurate, less error prone, more precisely defined and faster. For instance, with modern computers, it is possible to evaluate a complete free-to-trim, GZ curve, whilst simulating the shift of fluids in tanks in a couple of seconds thus making criteria evaluated from upright GM and traditional free surface moment correction to vertical centre of gravity obsolete. During the software development process, many stability criteria from many different organizations were reviewed. It is clear that the criteria of most regulatory bodies derive from the same original source, though there have often been different approaches taken to unit conversion and the derivation of constants. It was also extremely apparent, when translating criteria from the "plain English" of the stability documents to computer code, that many of the criteria are vague and open to different interpretation; sometimes, it even seemed that the intent of the criterion was different from the specification of the criterion. Computer code can only have one meaning -- a computer does exactly what it is told to do. Determining whether a criterion can be specified in computer code provides an excellent test of its unambiguity. Another observation was that there was no reference to accepted definitions of many common naval architecture terms; this can lead to difficulties and confusion where slight differences in accepted meanings for these terms vary from one country to another or from one naval architect to another.
@inproceedings{Couser2003, Use of computers in the design of high-speed craftPatrick CouserRINA at Ausmarine, Fremantle, Western Australia, October 2002. AbstractThe personal computer, as we know it, has only been around for perhaps 25 years, yet today they are completely taken for granted and pervade most aspect of our lives. The desktop computer of 2002 is as powerful as supercomputers of only 25 years ago. Due to the exponential growth in computing power, as observed by Gordon Moore in 1965, all ship designers have access to computational resources that would have been science fiction only ten years ago. This readily available computer power together with the maturation of design and analysis software has left designers in an unprecedented position to develop near optimum vessels. However, are we really making the most of this abundant computing resource? Is the fact that computers are now so widely accepted and readily accessible distracting us from thinking through problems clearly, making us focus on presentation rather than content and generally making us less efficient and preventing us from achieving our full potential? Who is the slave and who is the master?This paper reviews the development of computer hardware since the first modern computer approximately 50 years ago. It looks at future developments in both readily available computational power and new technologies and how these might impact on the design and operation of high-speed craft. It also looks at possible crossovers from other industries that could influence the way in which high-speed craft are designed and operated. Developments in design and computational methods and software and how these are linked to the available computational resource are also investigated. In particular the tools available for performance prediction and the state of the art in computational fluid dynamics (CFD) are examined and likely developments extrapolated. Will we ever see, the much talked about Holy Grail of CFD researchers, the "numerical towing tank"? The point at which new CFD codes become sufficiently mature to be used in routine design and a review of these analysis tools is also presented.
@inproceedings{Couser2002, Surface Fitting Using Genetic AlgorithmsAndrew MasonThe Naval Architect, The Royal Institution of Naval Architects, April 2002, pages 23-25. ExtractComputer aided hull design systems have been widely adopted and have dramatically reduced the amount of work needed to generate a fair linesplan. However, the rapid change in design technology from drawing board to computer has been responsible for the orphaning of an immense number of existing designs previously stored as lines drawings. Although it is possible to scan these drawings into a computer and create a 3D wireframe model relatively easily, creating a fair surface model suitable for analysis or further design modification is another matter ...
@inproceedings{Mason2002,
Seakeeping analysis for preliminary designPatrick CouserAusmarine 2000, Fremantle, Western Australia, November 2000. AbstractSeakeeping performance is becoming of increasing importance. With fast computers and sophisticated software readily available to designers, it is now possible for a vessel's seakeeping characteristics to be addressed much earlier in the design spiral.Regulatory bodies and operators are becoming increasingly aware of the importance of specifying seakeeping requirements which the vessel must meet. The software tools described in this paper will help designers address these issues. This paper gives a brief overview of what analysis techniques are available and gives a simple example of how the seakeeping performance of alternative designs may be compared.
@inproceedings{Couser2000, Towards predicting downwind performance of yachts in wavesDougal Harris, Giles Thomas and Martin RenilsonSTAB 2000, 7th International Conference on the Stability of Ships and Ocean Vehicles, Launceston, Tasmania, Australia, February 2000. AbstractThis paper reports on work conducted to date towards the development of a numerical simulation of yacht’s behaviour whilst sailing downwind in waves. Numerical models for the resistance, wave force and sail force have been incorporated into a longitudinal time domain simulation. The resistance and wave force components have been compared with experiments, conducted at the Australian Maritime Colleges' Ship Hydrodynamics Centre. The experiments used a series of IMS style yacht models designed by industry participants Murray Burns and Dovell. The effects of hull form parameters, appendages, encounter frequency and wave conditions on the wave force experienced by a yacht have been investigated and the results presented.
@inproceedings{Harris2000, Fourth Generation Software Tools for Workboat Design and ConstructionAndrew Mason and Philip ChristensenWork Boat World, Baird Publications, January 1999. IntroductionComputer Aided Design tools for Naval Architecture Design and Construction have undergone significant development in the last thirty years. Not only have they improved dramatically in capability, they have become more affordable as the cost of computers has reduced and become accessible to the designers and builders of even the smallest vessels.There is now a significant difference between the software used by large shipyards and that used by workboat designers and builders. This paper describes these differences and details why the approach to Computer Aided Design and Manufacture(CAD/CAM) adopted by workboat designers and builders needs to be fundamentally different from the methods adopted by large shipyards.
@article{MasonChristensen1999, Investigation into Wave Loads and CatamaransStephen M. Cook, Patrick Couser and Kim KlakaRINA Hydrodynamics of High Speed Craft Conference, London, UK, November 1999 AbstractCatamaran designs, whilst fundamentally unchanged from their historic predecessors, have developed rapidly over the last decade. As these vessels become larger and faster, accurate prediction of hull loads becomes increasingly important. Unlike more traditional vessels, high speed craft must have efficient, light-weight structures to maximise their payload and/or operating speed; however, the safety and structural integrity of the vessel must not be compromised.To obtain a better understanding of the behaviour and response of a catamaran in a seaway, an eight metre research catamaran, "Educat", has been built and instrumented with strain gauges and motion sensors. The research and findings to-date are presented in this paper, including: calibration of "Educat"; results of the sea trials and towing tank tests; and correlation with the numerical models.
@inproceedings{CookEtAl1999, Use of CFD techniques in the preliminary design of upwind sailsPatrick Couser and Norm DeaneThe 14th Chesapeake Sailing Yacht Symposium, Annapolis, Maryland USA, January 1999 AbstractThe results of the 1997 World Titles, held in Kingston, Canada, highlighted that there was considerable scope for improving the upwind performance of the International Mirror Class by making small adjustments, within the tolerances allowed by the class rule, to the sails and underwater foils. This paper describes some aspects of the Australian research and development programme in preparation for the 1999 World Titles to be held in South Africa in April.Computational methods, based on the vortex lattice method, have been used to provide direction and guidance for the on-the-water testing and trialing programme. The use of these theoretical tools has enabled a far wider range of sail, dagger board and rudder parameters to be investigated than would be possible using purely on-the-water testing. The usefulness of well-understood computational and numerical methods in sail and foil design has been demonstrated; it has also been shown that these tools are within the reach of relatively small budget research and development programmes. The proof of the pudding may be at the 1999 International Mirror Class World Titles... (watch this space)
@inproceedings{CouserDeane1999, An Improved Method for the Theoretical Prediction of the Wave Resistance of Transom-Stern Hulls Using a Slender Body ApproachPatrick Couser, A.F. Molland and J.F. WellicomeInternational Shipbuilding Progress, 45(444), December 1998. AbstractA catamaran comprises two demihulls and although the flow about the catamaran centre line is symmetric the flow about the centre lines of the individual demihulls is not. The asymmetric nature of the fluid crossflow around the demihulls causes sideforce and hence induced drag to be experienced on the demihulls. The sideforces generated by each demihull act in opposition and cancel whereas the induced drags of both demihulls act together to resist the forward motion of the vessel.The experimental procedures used to estimate the sideforce and induced drag are presented together with results for one hullform (5b) at two demihull separations (S/L=0.225 and S/L=0.329) and at several Froude Numbers. It has been shown that the induced drag generated by the demihulls is negligible despite the generation of significant sideforce. The sideforce produced was found to reduce rapidly with increasing demihull separation.
@article{CouserEtAl1998a, Experimental Measurement of Sideforce and Induced Drag on Catamaran DemihullsPatrick Couser, A.F. Molland and J.F. WellicomeInternational Shipbuilding Progress, 45(443), September 1998. AbstractAn existing numerical method for calculating the wave pattern and hence wave resistance of a body moving in a free surface has been improved and developed. A potential, non-lifting, model has been used with linearised free surface conditions to describe the flow around a body in a finite channel. The model has been developed specifically for calculating the wave pattern resistance of slender catamaran hulls with transom sterns, but may be applied to more general ship hull-forms provided that they have sufficiently high Length:Breadth ratio. The bodies are represented by planar arrays of sources on the local hull centre-lines. Particular attention has been made to the closure of the model in the region of the transom stern and results have been compared with experimental measurements carried out by the authors. Comparisons have also been made with results obtained from higher-order panel codes. Good correlation with both experiment and higher-order methods have been found for the slender hull forms investigated. The method presented offers orders of magnitude saving in computational effort over the higher-order methods without sacrificing accuracy.
@article{CouserEtAl1998b, Computational methods for investigating sail forces -- A case studyPatrick CouserYacht Vision'98, Auckland, New Zealand, January, 1998 AbstractComputational methods for calculating the flow of the wind over a yacht’s sails have been developed over many years. Amongst the pioneers in this field were Milgram (1968), Thrasher (1979) and Register (1983). These methods are under continuous development and refinement, for example Fiddes (1996).This paper presents the results of a case study using such a computational model. In this study the upwind performance of the International Mirror Class dinghy is investigated, with particular regard to the interaction between the jib and main sail and the effects of raking the mast aft. The latter point was of particular interest since the British sailors had adopted this rig set-up in preference to a vertical mast and were able to dominate the racing, particularly in lighter winds. The work presented here is part the Australian Maritime Engineering Cooperative Research Centre’s (AME CRC) Yacht Technology Research Program. This program involves the development of computational techniques for predicting yacht performance. Two of the principal tools are the velocity prediction program (VPP) and a vortex lattice model which is used to compute sail forces. After a presentation on the Yacht Technology Research Program at the recent Sailing Science Conference in Hobart, Tasmania (Couser 1997), the author was approached by Norm Deane2 and Steve Walker3 with the aim of using computational methods to investigate the aforementioned phenomenon. This work was undertaken since it contributed to the ongoing validation and improvement of these numerical methods and provided an ideal opportunity to apply theoretical sail analysis methods to a practical problem.
@inproceedings{Couser1998, Calm water powering predictions for high-speed catamaransP.R. Couser, A.F. Molland, N.A. Armstrong and I.K.A.P. UtamaFast'97, Sydney, NSW, Australia, July 1997 AbstractIn recent years, the rapid growth of the high-speed catamaran industry has highlighted problems in the way in which full scale vessel resistance is calculated from model tests. Using the traditional approach the total resistance can be broken down into Froude and Reynolds number dependent components. These are then scaled according to their respective scaling laws.The common practical method of separating the resistance components is by use of a form factor to estimate the total viscous component. In the case of the catamaran, the form factor will include the interaction effects between the demihulls: changes in the boundary layer due to the modified pressure field around the demihulls; and the influence on frictional resistance of the velocity augmentation between the demihulls. Of major interest, in the field of high-speed catamarans, is the appropriate magnitude of the form factor and its dependence on hull and operational parameters such as: length: displacement ratio; breadth:draught ratio; separation:length ratio and Froude number. Historically, the choice of form factor for slender catamaran forms has often been close to unity. However, recent research, from a number of independent researchers working in this field, has indicated that this may not be the case and that form factors greater than unity may be appropriate for these vessels. This paper presents the results of work carried out at the University of Southampton and the Australian Maritime Engineering Co-operative Research Centre which has sought to gain a better understanding of the resistance components of high-speed catamarans and the appropriate form factors to be used for resistance scaling.
@inproceedings{CouserEtA1997, |
Copyright © - Formation Design Systems Pty Ltd trading as FormSys 
pdf/382kB