中文摘要：

将径向基（RBF）神经网络应用到工程陶瓷缓进给大切深磨削领域,建立了磨削功率随砂轮速度、工作速度、磨削深度变化的预测模型。研究结果表明：预测值与实际值最大误差为5.30%,平均相对误差为3.2%,因此,径向基神经网络能准确地预测磨削功率的变化趋势。

英文摘要：

The application of radial basis function（RBF）neural network to engineering ceramics creep-feed has set up aprediction model for deep grinding field.According to the model,the grinding power changes as the grinding wheel speed,working speed and grinding depth change.Result shows that the maximum deviation between predicted value and real value is 5.30% and the average relative deviation is 3.2%.This means that the RBF neural network can accurately predict the change trend of grinding power.