Multidisciplinary Design Optimization For Mini Autonomous Underwater Vehicle

Mini autonomous underwater vehicle (AUV) is a typical complex system involvingresistance, structure, energy and propulsion disciplines. Traditional design method of AUVartificially separates these subjects which tends to reduce the overall performance of thesystem, so the introduction of multidisciplinary design optimization (MDO) theory to thedesign process of AUV will have great significance.Based on Isight software, researches of MDO techniques for cylindrical and flatunderwater vehicles were respectiely carried out. The main work is summarized as follows:(1) Each sub-discipline of the cylindrical underwater vehicle was studied. The cylindricalunderwater vehicle system is divided into four disciplines involving resistance, structure,energy and propulsion. Then detailed modeling and analysis of each subject were conducted,in which, approximation model technology applied to propulsion subject was introduced indetail. Based on the Sequential Quadratic Programming (NLPQL), Hooke-Jeeves DirectSearch Method (HJ) as well as Multi-island Genetic Algorithm (MIGA), each singlediscipline was optimized. When getting the optimization results, the most suitableoptimization algorithm for each subsystem was determined.(2) The application of MDO techniques to cylindrical underwater vehicle was researched.The interdisciplinary couplings of the four disciplines were studied, and then MDO modelsbased on Collaborative Optimization (CO) and Simultaneous Analysis and Design (AAO)methods were built respectively. Based on two MDO methods and and three optimizationalgorithms (NLPQL, HJ, MIGA), the multidisciplinary design optimization of cylindricalunderwater vehicle was conducted, and satisfactory results was gotten.(3) The application of MDO techniques to flat underwater vehicle was researched. Theflat underwater vehicle system is also divided into four disciplines involving resistance,structure, energy and propulsion. But the disciplinary analysis and modeling for flatunderwater vehicle is very different from cylindrical underwater vehicle. Then the same MDOmethods to cylindrical underwater vehicle were applied to the flat underwater vehicle andsatisfactory optimization results was gotten as well.(4) Comprehensive comparative analysis was made on these MDO results of twounderwater vehicles, and a set of method suitable for MDO of AUV was summed, that is the method of AAO based on NLPQL can be regarded as the first choice for multidisciplinarydesign optimization of mini autonomous underwater vehicle.