一种改进的遗传模拟退火算法及其应用

对遗传模拟退火算法进行了研究,首先对遗传算法的适应度线性尺度变换作了改进,然后在改进的遗传算法中融入了模拟退火算法,从而提高了遗传算法的全局寻优能力。以空心并联电抗器的优化为例,建立了空心并联电抗器的优化设计模型,给出了进行模拟退火操作时产生新解的邻域结构,采用改进的遗传模拟退火算法对空心并联电抗器进行优化设计,计算结果表明空心并联电抗器的损耗和重量分别减小了16.3%和11.5%,优化效果明显。

A Modified Genetic Simulated Annealing Algorithm and Its Application

Genetic simulated annealing algorithm is studied. Fitness linear scaling is modified firstly, and then the modified genetic algorithm is combined with the simulated annealing algorithm to improve the global convergence performance of genetic algorithm. The modified genetic simulated annealing algorithm is applied to the optimum design of air-core shunt reactor. The optimization model of air-core shunt reactor is presented. The neighborhood structure to generate new solution of simulated annealing operation is also proposed. The results show that the power loss and weight of air-core shunt reactor are decreased by 16.3% and 11.5% respectively