基于遗传算法优化的BP神经网络在粗糙度预测上的应用
Application of BP Neural Network with Genetic Algorithm to Roughness Prediction

针对BP神经网络预测工件表面粗糙度精度不高的问题，提出了一种基于遗传算法优化的BP神经网络预测方法。首先用遗传算法对BP神经网络的初始权值、阈值进行全局寻优，然后对优化的BP神经网络进行训练、预测。通过MATLAB进行了粗糙度预测仿真验证。结果表明：优化的BP神经网络比未优化的BP神经网络具有更高的预测精度。
A BP neural network prediction method of roughness with genetic algorithm (GA) is proposed to solve the problem of low prediction precision. In the present method, GA is firstly used to determine the initial weights and threshold by global optimization, and then the optimal BP neural network is used to train and predict the roughness. A simulation of the roughness prediction is executed with MATLAB. The simulation results show that the prediction accuracy of roughness with BP neural network is higher than that the un-optimal BP neural network