AbstractShape optimization is a growing field of interest in many areas of academic research, marine design,and manufacturing. As part of the Computational Research and Engineering Acquisition Tools andEnvironments Ships Hydromechanics Product, an effort is underway to develop a computationaltool set and process framework that can aid the ship designer in making informed decisions regard-ing the influence of the planned hull shape on its hydrodynamic characteristics, even at the earlieststages where decisions can have significant cost implications. Themajor goal of this effort is to utilizethe increasing experience gained in using these methods to assess shape optimization techniques andhow they might impact design for current and future naval ships. Additionally, this effort is aimed atestablishing an optimization framework within the bounds of a collaborative design environmentthat will result in improved performance and better understanding of preliminary ship designs at anearly stage. 58712
The initial effort demonstrated here is aimed at ship resistance, and examples are shownfor full ship and localized bow dome shaping related to the Joint High Speed Sealift hull concept.IntroductionAny ship design inherently involves optimiza-tion, as competing requirements and designparameters force the design to evolve, and asdesigners strive to deliver the most effective andefficient platform possible within the constraintsof time, budget, and performance requirements.A significant number of applications of compu-tational fluid dynamics (CFD) tools tohydrodynamic optimization, mostly for reduc-ing calm-water drag and wave patterns,demonstrate a growing interest in optimization.In addition, more recent ship design programswithin the US Navy illustrate some fundamentalchanges in mission and performance require-ments, and future ship designs may be radicallydifferent from current ships in the fleet. Onedifficulty with designing such new concepts isthe lack of experience from which to draw fromwhen performing design studies; thus, optimiza-tion techniques may be particularly useful.These issues point to a need for greater fidelity,robustness, and ease of use in the tools used inearly stage ship design.
The ComputationalResearch and Engineering Acquisition Tools andEnvironments (CREATE) program attempts toaddress this in its plan to develop and deploy setsof computational engineering design and analy-sis tools. It is expected that advances incomputers will allow for highly accurate designand analyses studies that can be carried outthroughout the design process. In order to eval-uate candidate designs and explore the designspace more thoroughly shape optimization is animportant component of the CREATE ShipsHydromechanics Product. The current programdevelopment plan includes fast parameterizedcodes to bound the design space and more accu-rate Reynolds-averaged Navier–Stokes (RANS)codes to better define the geometry and perfor-mance of the specified hull forms. The potentialfor hydrodynamic shape optimization has been demonstrated for a variety of different hullforms, including multihulls, in related efforts(see, e.g.,Wilson et al. 2009; Stern et al. 2007;Campana et al. 2006, 2009). The tools are basi-cally in place for performing hydrodynamicshape optimization of a hull form, but a signifi-cant effort is needed to demonstrate thiscapability for hull forms currently of interestto the Navy, to include the capability in auseable package and process, and to validate theprediction capability.The US Navy sees the need to change, at thehighest levels, and to take greater advantage ofexpanding computational capability. This wasrecently addressed in a memorandum fromCOMNAVSEA, establishing high-level capabil-ity goals for NAVSEA design synthesis andanalysis tools.1As part of preliminary and con-tract design it was stated that: ‘‘The goal forsynthesis and analysis tools used in this acquisi-tion phase is enabling the completion of a designiteration in 8 to 10 weeks including insight as tochanges needed for the next design iteration.’’RANS computations already have the capabilityto provide hydrodynamics predictions withinthis time frame. They are still too slow to pro-vide a comprehensive analysis of all aspects of aNavy ship design, which are needed; however,they can be used for studies of resistance, pow-ering, and maneuvering and by combining themwith optimization techniques can provide theneeded insight for design iterations.The goal of the CREATE program is to impactdesign in a meaningful way with high-fidelityhydrodynamic predictive tools. In order to dothis, it is necessary to obtain these codes into adesign environment. This means automating theuse of these codes as much as possible to allowfor running many hull variants that can providemeaningful ship behavior information to a shipdesigner. Shape optimization is a key componentof this effort, as well. Although the developmentof new software for future HPC resources is theplan for CREATE, there is also the near termCREATE goal of providing incremental capabil-ity and benefits throughout the CREATEprogramlifetime and part of that is using currentsoftware effectively. Consequently, a part of theCREATE Ship Hydromechanics Product isaimed at using existing codes for design studiesof relevant hulls forms and combining them withshape optimization algorithms to achieve betterperforming concepts. This also sets the stage forthe incorporation of the future CREATE high-end codes earlier into the design process as theybecome available. The higher-resolution physics-based RANS tools that are currently being usedwithin the Navy are showing phenomenalcapabilities for a wide range of geometries andconditions and address many of the hydrody-namic aspects of ship design. They are too slowfor early stage design, but they do provide amodeling framework for developing the nextgeneration of high resolution codes gearedtoward the next generation of computers, whichcan have an impact in the early stages of adesign. Validation is also still an issue that mustbe addressed and will receive considerableattention throughout the effort. There have beennumerous validation studies performed withthese codes, but not in a framework that has yetengendered confidence in the codes for design.Consequently, the current effort is also gearedtoward systematic validation over a range ofconditions for relevant hull forms. Ultimately,this will provide a design infrastructure toaddress many hull options.Current efforts have focused on use of thehydrodynamic analysis tools that are currentlyimplemented within the CREATE integratedhydrodynamic design environment (IHDE) orare planned for implementation in the future.This includes both linear (using slender shiptheory potential-based methods) and nonlinear(RANS) evaluations of hydrodynamic resistanceand comparison with experimental data. Inaddition, efforts have focused on developing anoptimization process that can be implementedwithin the IHDE framework. One of the keyelements that is necessary for integration with a1Memorandum from Admiral Sullivan on ‘‘Ship Designand Analysis Tools,’’ dated February 4, 2008. design environment is automation. To that end,an automated process has been implemented fordetermining the hull shape perturbations andevaluating the objective function for each per-turbed shape for linear methods. 源:自;优尔'-论.文,网·www.youerw.com/  船体初步设计中的船型优化英文文献和中文翻译:http://www.youerw.com/fanyi/lunwen_63696.html