Research Of Structure Topology Optimization Based On EFG And ESO

At present, the research of topo logy optimizatio n based on mesh-free method isstill under disc uss ion and it is far from we ll developed. But the inherent advantages ofmesh-free method make the idea of deve loping topolo gy optimizatio n based on itpromis ing. On the other hand, there are some shortcomings that need to be improvedfor improving the new topology optimization idea, such as, the instability, theineffic ienc y, and so on. In this paper, a newly deve loped topolo gy optimizationmethod called evo lutio nary structural optimization method is adopted to comb ine withmesh-free numeric calculatio n method to study the topology optimization oftwo-dimens iona l continuum structure. The ma in conte nts of this paper are as fo llows:1. Using the “ANSYS Parametric Design Language” of ANSYS software, someANSYS program codes used for simulate the evolutio nary topology optimizationprocess of two-dime ns iona l continuum models are written. The ranges of para metersof evolutionary structura l optimizatio n method are decided by repeated test.2. The procedure step of topology optimizatio n based on e leme nt-free Galerk inmethod and evolutionary structura l optimizatio n method is give n on the basis ofprevious work. The influence to the structure a nalys is that comes fro m continua lnodes’ deletion during the topolo gy optimization process is discussed. To tackle withthe influe nce, three possib le ways are provided in this paper, they are:1、 letting thedeleted nodes comp lete ly not contrib ute the structura l ana lys is;2、tackling with theshape functio n of deleted nodes directly;3、 us ing “node birth and death” technolo gy.For each way, some MATLAB progra m codes that used for exa mining the valid ity oftopology optimization based on e leme nt-free Ga lerkin method and evo lutio narystructura l optimization method are writte n. The results show that the second and thethird idea are effective. And the second way is the best.3. Parameter such as node density, the size of influe nce doma in, etc. that affectsthe results o f topology optimization based on ele me nt-free optimization method andevolutionary structura l optimizatio n method are discussed. Recommendations aremade to he lp choose the value of these parameters.4. The princip le of topology optimization based on variable dens ity method andele ment-free method is introduced. A co mparison is made between two topologyoptimizatio n methods in this paper. The results show that the topolo gy optimizationbased on e le ment-free Ga lerkin method and evo lutio nary structura l method can increase efficie ncy o f topolo gy optimizat ion based on mes h-free method effective ly.In summary, in this paper, the ele ment-free Galerkin method and the evolutionarystructura l optimization method is successfully co mbined to solve the topologyoptimizatio n proble m o f two-dime ns iona l continuum model. The results s how that itcan increase effic ie ncy effective ly. And this provides a new way of topologyoptimizatio n based on mesh-free method.