Handling variable string lengths in GA based structural topology optimization

摘要: Binary coded genetic algorithms (GA’s) have been used effectively in topological design of discrete structural systems. In a majority such applications, the topology is extracted from pre-defined ‘structural universe’, set all permissible joints and elements that can be development optimal design. presence dense GA search process must contend with very long string lengths, attendant degradation effectiveness process. The present paper presents novel approach for handling variable lengths GA-based Varying population requires redefinition crossover process, both inter- intra-species mechanisms are explored paper. use micro-GA’s proposed as an to increasing efficiency problems involving large number candidate topologies. strategies implemented representative algebraic problems, truss design, layout stiffened composite panel.