基于工艺流程的零件成组方法及其实现

提出了基于遗传算法和BP网络的零件成组方法.首先,建立零件聚类的数学模型,给出了一种基于遗传算法的求解方法,使用整数编码构造的染色体和合适的遗传算子来满足聚类模型复杂约束的要求,并利用VC编程实现该算法.其次,将一批已聚类成组的零件作为训练样本,利用Matlab中神经网络的工具箱对BP网络进行训练,通过训练后的网络对新的零件进行仿真,从而实现新零件的匹配.最后通过实例证实了该方法的可行性和优越性.

Method and realization of parts-grouping based on technical process

The parts grouping method that based on genetic algorithm and BP neural network is proposed.At first,the grouping model is built,and a solving method based on the genetic algorithm is given,which uses integer code building the chromosome and proper genetic operator to satisfy the requirement of group model,and is implemented by VC.Second,some parts which have been grouped are used as the sample to train the BP neural network with the toolbox in Matlab,then simulate a new part on the network,and it works out that the new part belonged to which group.At last,a case study shows that the method is feasible and ascendant.