Study On Optimization Method Of Ship Resistance Performance Based On RANS Method And SBD Technology

The prediction method of ship resistance performance and the optimization method of hull form design and flotation design are very important in the optimization of ship resistance performance.For conventional drainage ships,a good design of hull form is the important foundation among all drag reduction methods;for high-speed planing vessels,a good flotation design is an important means to balance the relationship between high speed with low resistance and the stability of longitudinal motion.With the continuous improvement of computer performance and the development of numerical simulation technology,the prediction method of ship resistance performance based on RANS method has been widely studied and developed;At the same time,in order to reduce the computational cost in the optimization,SBD technology has emerged,and has been developed and applied in many fields such as ships,vehicles,aerospaces and so on.At present,in the prediction of ship resistance performance based on RANS method,the setting of boundary layer often require excessive validation works,and lack of simple setting suggestions;The size of computational domain and the selection of turbulence model are mostly based on experience,and lack of reasonable reference.At the same time,when SBD technology is applied in the field of ship design optimization,the comparative discussion on the approximate modeling methods is insufficient,and lack of improvement for specific problems.Therefore,in allusion to these problems,this paper starts with two aspects of the prediction method of ship resistance performance based on RANS method and the optimization method of resistance performance based on SBD technology.Firstly,in the calculation based on RANS method,the setting of boundary layer grid,the design of computational domain size and the selection of turbulence models for two different ship types,i.e.conventional drainage ships and planing vessels,are discussed;then,the modeling accuracy of different approximate modeling methods is compared and analyzed,and corresponding improvements are proposed for two specific optimization problems.Specifically,there are four aspects in this paper:(1)Taking two different ship types of KCS model and Fridsma planing vessel as calculation objects,the prediction of ship resistance performance and longitudinal motion based on RANS method is carried out.The results show that the errors between the prediction results and the towing tank experimental data are small,which verifies the rationality of the boundary layer setting method and the computational domain size design method proposed in this paper,and recommendations for the selection of turbulence models for this two ship types are given.(2)In the research of resistance performance optimization method based on hull form design,the SBD optimization technology,which combines the OLHS sampling method,Kriging modeling method,sensitivity analysis and NSGA-?optimization algorithm,is applied to the stern optimization of a twin-skeg ocean going fishing vessel.The process of optimization design with SBD technology is completely demonstrated,and the accuracy of different approximate modeling methods is compared and analyzed.Finally,the optimal design of skeg with average drag reduction of 5.4%under four speeds and 5.6%at design speed is obtained,which reflects the reliability of SBD optimization technology.(3)In the research of resistance performance optimization method of low-speed leadgebow ship based on principle dimensions design,due to the design variables in the geometric modeling and the design variables in the approximate modeling are different variables with coupling relationship,resulting in the "gap" at the edge area of the design space during the data training.In this paper,a Kriging method based on the edge correction samples is proposed,according to the gap distance evenly generates supplementary sample points in edge area,to ensure that Kriging approximate model can cover the whole design space completely.After verifying that the approximate model based on ECS-Kriging method has better approximate accuracy at the edge area of design space,it is applied to the principle dimensions optimization of KVLCC2 oil tanker,KCS container ship and 66000DWT bulk carrier,and optimal principal dimensions design with drag reduction of 7.85%,5.08%and 8.24%respectively are obtained.(4)In the research of resistance performance optimization method of planing vessels based on flotation design,due to the difficulty of convergence of numerical calculation reflects the stability of longitudinal motion of planing vessels,then it affects the distance between the sample point and the navigation state boundary,it is unable to simply rely on the basic SVM algorithm to build the boundary between the sample points of stable planing state and longitudinal motion instability.In this paper,a SVM algorithm based on deviation degree constraint is proposed to solve this problem.The convergence difficulty of sample points is added to SVM algorithm as deviation degree constraints,and the flotation optimization design method considering "Porpoising Motion" is studied.Taking the flotation optimization design of a V-type high-speed planing vessel as an example,by reasonable fitting the boundary between stable planing state and longitudinal motion instability,the flotation optimization design of 5.15%drag reduction at design speed is obtained.In this paper,based on the research background of ship resistance performance optimization design,some details of prediction of ship resistance performance and longitudinal motion based on RANS method are discussed,the SBD technology for ship hull form design,principle dimensions design and flotation design is proposed.Hopefully,the research of this paper will contribute to the development of ship resistance performance optimization method.