基于改进遗传算法的孔群数控加工路径优化

Research on Path Optimization of Hole Group Machining Based on Improved Genetic Algorithm

针对孔群加工路径优化中遗传算法存在的局部最优和收敛速度慢等问题,提出采用模拟退火算法改进遗传算法进行路径优化。首先根据孔群数控加工的特点建立数学模型,采用遗传算法设计种群编码,建立适应度函数选择优秀种群,并对保留的优秀种群进行交叉、变异等操作,实现种群进化,其次引入模拟退火算法对其适应度函数进行拉伸处理,调整种群进化差异性而加速寻优进度,同时采用改进的Metropolis准则调整接受概率,调节旧种群和新种群的进化程度,增强遗传算法的全局搜索能力。实例表明:改进算法用于某模具的孔群加工,有效克服遗传算法的早熟现象,缩短收敛次数,平均路径缩短比例达6.9%,提高了加工效率,效果良好。

Aiming at the problems of local optimum and slow convergence of genetic algorithm in the optimization of hole group machining path, an improved genetic algorithm based on simulated annealing algorithm is proposed. Firstly, mathematical model is established according to the characteristics of hole group machining, then the encode model of the population is designed based on genetic algorithm, the operations of crossover and mutation of outstanding population retention are implemented by the fitness function to make the population evolve. Secondly, the genetic algorithm is combined with simulated annealing algorithm to draw the fitness function for adjusting the evolution of population differences and accelerating the optimization progress, at the same time the improved Metropolis standards are used to adjust the probability of acceptance and evolution degree of the old and new regulation of population for enhancing the global search ability of genetic algorithm. Examples show that the algorithm has been used in hole group machining of the mould. The improved genetic algorithm compared with traditional genetic algorithm can overcome the prematurity of traditional genetic algorithm effectively, reduce the number of convergence and reduce by more than 6.9% on average path, which has increased of efficiency and has good effect.