基于GA-BP神经网络的机床主轴刚度预测研究

Research on Prediction of Machine Tool Spindle Stiffness Based on GA-BP Neural Network

提出一种基于遗传算法(GA)优化的BP神经网络的机床主轴刚度预测模型,以主轴悬伸量、前后轴承间距、主轴当量外径、前轴承径向刚度、后轴承径向刚度为输入,以主轴末端刚度为输出,训练神经网络,可以预测主轴刚度。研究表明,经过遗传算法优化的BP神经网络模型较未经遗传算法优化的BP神经网络模型而言,拥有较高的预测精度。

A prediction model of machine tool spindle stiffness based on BP neural network optimized by genetic algorithm(GA)is proposed. The main spindle stiffness can be predicted by training neural network. The main spindle overhang,front and rear bearing spacing, equivalent outer diameter of main spindle, radial stiffness of front and rear bearing are taken as input, and the end stiffness of main spindle is taken as output.The research shows that the BP neural network model optimized by genetic algorithm has higher prediction accuracy than the BP neural network model that are not optimized by genetic algorithm.