基于零件重要特征参数编码的RBF神经网络在成组技术(GT)中的应用研究

Study on the Application of RBF Neural Network in Group Technology Based on the Encoding of the Important Feature Parameters of Components

成组技术实现的关键在于严格、科学地判断各类零件的相似程度,利用分类系统准确地将零件分类成组。笔者研究了利用RBF神经网络实现零件分类的方法,并找到了一种易于工程实践的零件成组技术。

The key in the realization of group technology lies in the rigid, scientific judgment of the degree of the similarity of different components and the utilization of classifying system to group components precisely. This paper studies the method to classify components by means of RBF neural network, and works out a technology of component grouping convenient for engineering practice.