自适应遗传退火算法的圆形件下料问题求解

Solving circular blanking problem with adaptive genetic annealing algorithm

针对长度不限、宽度固定的卷材的直冲圆形件下料问题,采用自适应遗传模拟退火算法(adaptive genetic simulatecl annearling algorithm,AGSA)优化毛坯序列,并采用最佳位置(best location position,BLP)算法决定毛坯放置位置。在遗传算法的基础上,引入环形交叉策略和自适应交叉变异概率,有效地提高收敛速度。将模拟退火算法与遗传算法结合,通过退温机制更改接受概率,避免遗传算法陷入局部最优解导致早熟。实验结果表明:本文提出的算法计算时间合理,能有效提高收敛速度和材料利用率。

Aiming at the problem of circular stock problem on infinite length and fixed width coils,in this paper,adaptive genetic simulated annealing algorithm is used to optimize the items sequence,and the Best Location Position algorithm is used to determine the blanking position. Based on genetic algorithm,the introduction of ring-shaped crossover strategy and adaptive crossover mutation probability can improve the convergence rate effectively. By applying temperature retreat mechanism to change acceptance probability,the combination of simulated annealing algorithm and genetic algorithm can avoid the local optimal solution and precocious. Experimental result proves that the computational time is reasonable,and the presented algorithm can improve the coverage rate and average utilization rate of material effectively.