基于有限元及神经网络的磨削温度仿真预测

为了控制由于磨削温度过高引起的工件表面热损伤,对磨削温度场的分布进行了仿真和预测研究。首先,采用有限元法对磨削温度场的分布状况进行了仿真研究,并通过实验验证了仿真结果的准确性;随后,以仿真结果作为训练样本采用BP神经网络对不同条件下的磨削温度进行了预测,通过与仿真结果的比较验证了BP神经网络预测模型的准确性。结果表明：采用有限元和神经网络相结合的方法对磨削温度进行仿真预测具有较高的准确性,为实际应用中磨削参数的选取提供了理论依据。

Simulation and forecast of the grinding temperature based on finite element and neural network

In order to control the fire damage of the workpiece resulted from the higher grinding temperature, the distribution of the grinding temperature filed was simulated and forecasted. Firstly, the grinding temperature flied was simulated by FEM method, and the accuracy of the simulation result was verified by experiment. Then, the grinding temperatures of different conditions were forecasted by BP neural network with the simulation results as training samples, and the veracity of the forecast model was proved though the comparison between the results of the prediction and simulation. The results indicate that the simulation and forecast of the grinding temperature based on finite element and neural network method is accurate, and it can provide theory base for the choosing of grinding parameters in implementation.