摘要
针对多目标产品配置方案寻优问题,构建了性能、价格、交货期的优化配置数学模型,考虑配置模块的互斥互补约束、个性化约束,采用基于改进差分元胞遗传算法求解最优配置。该算法根据邻居非支配解个体情况,选择不同的差分交叉策略,同时使用基于信息熵的优势个体评价准则更新中心个体,提高该算法的全局寻优能力和收敛速度。最后通过相关算例对比,验证改进算法的可行性并用于实际配置问题求解。
This paper build a mathematic model of performance,price,and delivery time for the optimal solutions of multi-objective product configuration. In viewof the incompatible and complementary constraints,personalized constraints,the optimal product configuration which used to improve the differential evolution cellular genetic algorithm is adopted. According to the neighbor non dominated solutions,the algorithm chooses different differential crossover strategies. At the same time,central individuals are updated by using the rule of advanced individual evaluation which is based on information entropy,then the global optimization ability and convergence speed of the algorithm are improved. At last,this paper verifies the feasibility of the improved algorithm by comparison with relevant examples,and applies it to the actual configuration problem solving.
出处
《组合机床与自动化加工技术》
北大核心
2017年第3期60-63,共4页
Modular Machine Tool & Automatic Manufacturing Technique
关键词
产品配置
差分元胞遗传算法
多目标
product configuration
differential evolution cellular genetic algorithm
multi-objective
