구조 최적화에서 적응 시뮬레이티드 애닐링의 냉각변수에 대한 연구

The increase of computing power makes stochastic optimization algorithms available in structural design. One of the stochastic algorithms, simulated annealing algorithm, has been applied to various structural optimization problems. By applying several cooling schedules such as simulated annealing (SA), Boltzmann annealing (BA), fast annealing (FA) and adaptive simulated annealing (ASA), truss structures are optimized to improve the quality of objective functions and reduce the number of function evaluations. In this paper, many cooling parameters have been applied to the cooling schedule of ASA. The influence of cooling parameters is investigated to find the rules of thumb for using ASA. In addition, the cooling schedule combined with BA and ASA is applied to the optimization of ten bar-truss and twenty five bar-truss structure.