Font Size: a A A

Topology Optimization Theory Based On Genetic Algorithm And Its Application In Micro Satellite Design

Posted on:2018-04-26Degree:MasterType:Thesis
Country:ChinaCandidate:H XuFull Text:PDF
GTID:2392330623450757Subject:Aeronautical and Astronautical Science and Technology
Abstract/Summary:PDF Full Text Request
Micro-satellites have the advantages of low cost,short development cycle.The topology optimization method can consider the optimal material distribution in the early stage,which is conducive to the size and shape optimization at the later stage,reducing the design cost and the development cycle.Therefore,the topology optimization is an important research direction of micro-satellite structure design.In this paper,micro-satellite structure topology optimization method and application are studied based on genetic algorithms.The method of objective function design with punishment strategy is proposed.The analysis of topology connectivity and the handling method of unit hinges were established.The Elitist Genetic Algorithm(EGA)is used to calculate the topology optimization problem of plane cantilever beam and simply supported beams.The results show that the EGA optimization results do not exist in the gray area.Compared with the traditional SIMP method,the calculation efficiency of EGA is relatively low.In addition,its smallest degree of compliance is poor.Aiming at the precocity of EGA,multi-population genetic algorithm(MPGA)is used to solve the problem of plane cantilever beam and simply supported beam.The results show that MPGA is a good solution to the premature phenomenon and the global optimal solution is better than the SIMP method.However,at the same time,the MPGA still has shortcomings such as more design variables and slower convergence rate.Aiming at the defect of MPGA,a design space reduction strategy based on approximate density distribution(ADD)is proposed.By using the density of a given sampling point and approximating the finite element density by surface spline interpolation,this method can only select the density interpolation of a few points to approximate the density of a large number of finite elements,greatly reducing the number of design variables.Combined with MPGA,ADD-MPGA method is proposed.Then SIMP-ADD-MPGA method is proposed lately.This method can eliminate some invalid cells at the beginning of optimization and reduce the design space.Numerical examples demonstrate the effectiveness of the above method.The SIMP-ADD-MPGA method is applied to the structural design of micro-satellites.The topology optimization of support leg of "TianTuo 3" satellite is carried out.The topology with the highest stiffness under the given conditions is obtained.Based on this,further size and shape optimization is carried out,resulting in a 17.93% reduction in mass and a 4.52% reduction in the axial fundamental frequency.The method proposed by the paper and the results of application examples can provide valuable references for topology optimization design of microsatellite structures.
Keywords/Search Tags:Genetic Algorithm, Topology Optimization, Elitist Genetic Algorithm, Approximate Density Distribution, SIMP method
PDF Full Text Request
Related items