Research On The Continuum Structure Topology Optimization Based On Improved Genetic Algorithm

Structure optimization plays an increasingly important role in the field of engineering and technology. This paper has lucubrated the higher levels topology optimization of structure optimization. Aimed at some deficiencies in the topology optimization method at present, for example, easily fall into the local optimal solution, low quality to optimize the structure, low efficiency to solve large-scale structure and so on, this paper has carried on deeper analysis,research and improved the traditional structural topology optimization method. Finally, it is improved high effect of the structure optimization by calculation.Firstly, this article combining the improved Genetic Algorithm (GA) with Bi-directional Evolutionary Structural Optimization method (BESO), puts forward the Bi-directional Evolutionary Structural Optimization method based on Improved Genetic Algorithm(IGA-BESO). The proposed method not only has solved the problem of the BESO method which is easy to get into the local optimal solution when optimizes the structure, but also has solved the problem of low efficiency and complexity of the GA algorithm when optimizes the large-scale structure. The improved crossover and mutation operation with penalty factor solves the problem of unit disconnection phenomenon which is liable to occur in optimization process.Secondly,this paper expands to optimize the structure with multiple load condition constraint through the IGA-BESO method. For dealing well with the relationship between each working condition, this article builds the objective function through the method of linear weighted average objective function of each working condition in structural optimization.This paper determines the weights of each working conditions through the α-method,and avoids the phenomenon of "sick load" effectively through the hierarchical optimization technology.At the same time, this paper expands to optimize the structure with periodic constraint through the IGA-BESO method. In this paper, in order to satisfy the periodic constraints, the whole structure is divided into the form of unit cell. It optimizes the representative unit cell of the structure in the process of optimization,and synchronously updates other unit cells,which can not only simplify the optimization model, but also reduce the amount of calculation and improve the optimization efficiency.Finally, this paper has optimized several 2D and 3D typical structures in practice engineering with IGA-BESO method, it confirms the reliability, superiority and versatility of the IGA-BESO method through sloving the problem of single working load, multiple working load constraints and the periodic constraints structure optimization.