基于RBF神经网络的产品概念设计方案评价

Product conceptual design schemes evaluation based on RBF neural network

分析了现有评价方法存在的问题,利用Matlab神经网络工具箱构建了RBF网络模型,并以冰箱为实例进行评价。RBF神经网络采用监督学习算法和正交最小平方(OLS)算法决定基函数的中心、方差以及隐含层到输出层的权值。与BP神经网络模型的评价结果对比,建立的RBF神经网络评价模型具有更高的预测精度,收敛速度更快。

The problems of existing evaluation methods are analyzed. The RBF neural network model is constructed by using Matlab program, and is applied to refrigerator evaluation. Supervision learning algorithm and the orthogonal least-squares （OLS） algorithm arc adopted by RBF neural network to decide the center of basic function, variance and weights between hidden layer and output layer. Compared with the results by using BP neural network, the evaluation model constructed by RBF can speed up convergence with higher precision.