Low-cost Unmanned Underwater Experimentation Flat Optimization Design Research

Recently,with the research of marine resources development and naval operations,the demands of experimentation flat that can carry experimentation equipment are increased.As a flexible,low-cost and facilitated installation new underwater equipment,UUV(Unmanned Underwater Vehicle)can satisfied the experiment well.This paper aim to optimize design an experimentation flat for provided experimentation equipment.Primary target is the research about sequential quadratic programming,genetic algorithm and simulated annealing algorithm,combinatorial optimization of UUV ’ s different single system and Multidisciplinary Design and Optimization of UUV.The main research contents are as follows:Firstly,start with the optimization algorithm of UUV.Do the research about the mathematics foundation and calculation process of sequential quadratic programming,genetic algorithm and simulated annealing algorithm.Finish the modeling of UUV including the bone structure,battery,compressive cabin and steering engine based on experimentation equipment.Then,analyze and research different kinds of MDO(Multidisciplinary Design and Optimization).Researched six different MDO methods and list the calculation process of them.Brief analysis every MDO’s computing characteristics,put particular emphasis on CO(Collaborative Optimization)and AAO(Simultaneous Analysis and Design).Next,optimization analysis every single system of UUV.Divides UUV into resistance,structure,energy and propulsion four independent single system.Modeling and optimization analysis every single system based on isight.Select NLPQL(Sequential Quadratic Programming)combined with MIGA(Multi-Island Genetic Algorithm)and NLPQL combined with ASA(Adaptive Simulated Annealing)as Combinatorial Optimization method to optimization analysis single system,and get a better Combinatorial Optimization method.Finally,execute the research of MDO of UUV.After realized interdisciplinary couplings between four systems,build the model of CO and AAO.Do the MDO to UUV with NLPQL combined MIGA and NLPQL combined ASA.At last get a better optimization result.