基于免疫遗传算法的不规则件排样优化问题求解

基于遗传算法难以保持群体的多样性及存在易早熟、效率低的缺陷,提出免疫遗传算法应用于不规则零件排样的优化方法。该算法在遗传算法的全局随机搜索基础上,借鉴了人工免疫系统中的免疫记忆和浓度机制。通过疫苗接种实现种群个体中基因位的局部调整优化,并将其优良个体保存于免疫记忆库中,提高了算法的搜索速度。同时浓度机制保证了遗传交叉和变异过程中生成下代种群个体的多样性,扩大了搜索空间,更利于最优解的获取。该方法在开发的不规则件排样系统中进行了实算求解,通过与标准遗传算法的实验结果比对,板材的利用效率得到显著提高。

The Solution for Irregular Parts Nesting Problem Based on Immune Genetic Algorithm

A novel solution for 2D irregular parts nesting with immune genetic algorithm(IGA) was presented,which overcome the shortages of premature constringency and low efficiency existing in genetic algorithms(GA).The immune memory and concentration mechanism of artificial immune system was introduced in global random searching of IGA.Vaccination realized individual gene a local adjustment and optimization,and the best individual could be saved in immune memory library to improve the search speed of algorithm.At the same time the concentration mechanism ensured genetic population diversity during crossover and mutation process,expanded the searching space,more conducive to the optimal solution of the acquisition.Comparing with the standard genetic algorithm,the experimental results by immune genetic algorithm showed that the utilization of material is increased,the effectiveness has been fairly proved in solving irregular parts nesting problem.