基于COIN算法的工艺原则布置优化  

Optimization of Process Principle Arrangement Based on Coincidence Algorithm

作　　者：林巨广[1] 荣海龙[1] 汪洋[1] 

机构地区：[1]合肥工业大学机械与汽车工程学院,合肥230009

出　　处：《组合机床与自动化加工技术》2016年第11期138-140,144,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家支撑项目:多平台高节拍汽车车身柔性焊装自动化生产线(2012BAF06B01);国家智能制造装备发展专项项目(发改办高科[2011]2548号)

摘　　要：为了解决焊装工厂的工艺原则布置问题,文章运用COIN算法(Coincidence Algorithm),用联合概率矩阵模型代替传统遗传算法的交叉和变异,来更新产生新的子代种群,并在学习和优化过程中利用负相关学习与传统的正相关学习相结合的方法,保持了种群的多样性,提高了算法的收敛速度。最后通过运用该算法解决焊装工厂的工艺原则布置问题,并以遗传算法做对比,验证了COIN算法的有效性。To solve the problem of process principle arrangement,the Coincidence Algorithm which replaces the traditional genetic algorithm crossover and mutation with the joint probability matrix model uses both good and not-good solutions that is defined by fitness to generate the population in this paper. It maintains the diversity of the population and improves the convergence rate. Finally,using the Coincidence Algorithm to solve the problem of process principle arrangement,it verifies the validity of Coincidence Algorithm com-pared with Genetic Algorithms.

关 键 词：COIN算法 遗传算法 工艺原则布置 

分 类 号：TH162[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]