Energy Efficiency Optimization Of Autonomous Underwater Vehicle Based On CFD Approximate Model

With the continuous development of science and technology and the increasing depletion of resources on the land,more and more people pay attention to the huge and abundant marine resources.As an important tool for the exploration of marine resources,the AUV(Autonomous Underwater Vehicle)has an irreplaceable role.Since the AUV can not be connected to the mother ship in underwater operations,it is necessary to carry as much energy as possible in a limited space and perform as long as possible.Modern exploration missions require further search distances and longer detection times,which requires the AUV to be energy efficient.In such an optimization system of the AUV,the primary factor is the direct performance of hull resistance.Therefore,the optimization of main hull resistance is the most direct and effective energy saving method.That is to say,lines to minimize energy dissipation has a higher theoretical value and practical significance.The research on drag optimization can not be separated from the support of CFD(computational fluid dynamic)and intelligent optimization algorithm.However,the CFD calculation of hull resistance also requires a certain period of time,intelligent optimization algorithms need to search a large number of iterations in the selected space to find the optimal solution.Therefore,it is necessary to introduce an approximate 1 model which takes less time to replace the CFD calculation,and finally get the optimal energy efficient lines by approximate model.Firstly,the advantages and disadvantages of the three drag calculation methods are compared,which are empirical formula,CFD calculation and model test.The validity of 2D unstructured grid method is verified by comparing 3D structured grids method.The results of two groups of empirical formulas,CFD calculation and model test are compared to verify the accuracy of the CFD algorithm.Secondly,the theory and method of the experimental design and three kinds of approximate models are studied,and an automatic approximate model integration platform is established,which is composed of the generation of hull type,grid and CFD resistance.Comparison of RSM model,RBF model and Kriging model,analysis of main parameters of the response law and error results.The four order RSM model of the speed of four groups is established,and all the models are weighted to get a fitting mod-el.The fitting model can be applied to the drag optimization of multi speed.Finally,the intelligent optimization platform is built,and three kinds of intelligent optimization algorithms are used to optimize the minimum drag and the maximum displacement.Comparison of MIGA algorithm,ASA algorithm and MOPSO algorithm,evaluate of the results and reasons for the optimization.The MIGA algorithm is used to optimize the minimum drag under fixed displacement and fixed LCB,and the optimal fitting lines and 1.5m/s speed lines are obtained.At last,the characteristics and changing rules of the two optimal lines are analyzed.