摘要
对遗传算法特别是自适应遗传算法作了分析,重点研究了交叉算子和变异算子对遗传算法收敛性的影响,提出了一种改进的自适应遗传算子的方法.该方法可在遗传模式得到保证的基础上加快新个体的产生速度,所构造的遗传算子随适应值自动变化,对远离最优值的个体采用较大遗传算子值,对接近最优值的个体采用较小遗传算子值,以提高得到全局最优解的概率.通过测试函数的求解,验证了所构造的自适应算子的有效性和正确性.实算结果表明,在无刷直流电机的优化设计中,改进后的自适应遗传算法可在满足各项性能指标的前提下取得良好的优化效果,得到全局最优解的概率较改进前有明显提高.
A method using genetic algorithm (GA) is presented for the design optimization of brushless direct current (BLDC) motor. In order to improve the convergence of GA, new crossover operator and mutation operator are adjusted according to the fitness values. The larger the difference between the parents and the optimum fitness is, the larger the values of crossover and mutation are, and the operation is performed along with the convergence direction rapidly. The smaller the difference between the parents and the optimum fitness is, the smaller the values of crossover and mutation are, and the operation is not located at the local optimum value. An improved algorithm is proposed and successfully applied to the design optimization of BLDC motor. The improved algorithm is compared with those of other methods, and the example shows that the optimization according to the proposed algorithm can obtain satisfactory results with better performance, lower price and shorter computation time.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2002年第12期1215-1218,共4页
Journal of Xi'an Jiaotong University
