基于卷积神经网络的平面磨削温度预测  

Surface Grinding Temperature Prediction Based on Convolutional Neural Network

作　　者：孙为钊 周俊[1] SUN Wei-Zhao;ZHOU Jun(School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)

机构地区：[1]上海工程技术大学机械与汽车工程学院

出　　处：《计算机系统应用》2020年第2期244-249,共6页Computer Systems & Applications

摘　　要：为了减少磨削温度过高给零件带来热损伤等的负面影响,并提高零件产量、质量,本文建立了基于卷积神经网络的平面磨削温度预测模型.首先通过有限元仿真获得温度数据,并进行预处理,然后利用Google开源深度学习工具TensorFlow编写卷积神经网络程序,最后得到预测结果并与仿真值进行比较.结果表明,本文提出的基于卷积神经网络的磨削温度预测模型具有很强的学习能力以及非线性拟合能力,大大提高了磨削温度预测精度.In order to reduce the negative impact of excessive grinding temperature on the thermal damage of parts,and to improve the yield and quality of parts,this study establishes a surface grinding temperature prediction model based on convolutional neural network.Firstly,the temperature data is obtained through finite element simulation,and preprocessing is performed.Then,the convolutional neural network program is written by Google’s open-end learning tool TensorFlow,and finally the prediction result is obtained and compared with the simulation value.The results show that the grinding temperature prediction model based on convolutional neural network has strong learning ability and nonlinear fitting ability,which greatly improves the prediction accuracy of grinding temperature.

关 键 词：卷积神经网络 磨削温度 深度学习 人工神经网络 机器学习 

分 类 号：TG580.14[金属学及工艺—金属切削加工及机床] TP183[自动化与计算机技术—控制理论与控制工程]