GA-BP神经网络在抛光材料去除率预测中的应用  

Application of GA-BP neural network for the prediction of polishing material removal rate

作　　者：吴瑞喜 柳奇搏 张善青[1,2] 王桂莲[1,2] WU Ruixi;LIU Qibo;ZHANG Shanqing;WANG Guilian(Tianjin Key Laboratory for Advanced Mechatronic System Design and Intelligent Control,Tianjin University of Technology,Tianjin 300384,China;National Demonstration Center for Experimental Mechanical and Electrical Engineering Education,Tianjin University of Technology,Tianjin 300384,China)

机构地区：[1]天津理工大学天津市先进机电系统设计与智能控制重点实验室,天津300384 [2]天津理工大学机电工程国家级实验教学示范中心,天津300384

出　　处：《天津理工大学学报》2024年第4期16-22,共7页Journal of Tianjin University of Technology

基　　金：国家重点研发计划项目(2018YFB1308900)。

摘　　要：材料去除率(material removal rate,MRR)是抛光过程中调控工艺参数和评估加工效果的一项重要指标。为了有效地预测抛光过程中工件的材料去除率,建立了一种运用遗传算法(genetic algorithm,GA)与反向传播(back propagation,BP)神经网络相结合的MRR预测模型,利用GA的全局寻优能力对BP神经网络的初始权重和阈值进行优化。通过随机森林(random forest,RF)算法选择出对MRR预测结果影响显著的参数变量作为神经网络的输入,运用有关实验数据对网络模型进行训练与预测分析,并与同等条件下运用传统BP神经网络建立的预测模型进行对比分析。结果表明:GA-BP神经网络具有较高的预测精度。Material removal rate(MRR)is an important indicator of adjusting process parameters and evaluating processing effect during polishing.In order to effectively predict the material removal rate of the workpiece in the polishing process,the MRR prediction model using genetic algorithm(GA)combined with BP neural network is established,and the initial weights and thresholds of the BP neural network are optimized by using the global optimization-seeking capability of GA.The random forest algorithm(RF)is adopted to select the parameter variables with significant influence on the MRR prediction results as the input of the neural network,and the network model is trained and predicted by using the relevant experimental data,and compared with the prediction model established by traditional BP neural network under the same conditions,the results show that GA-BP neural network has higher prediction accuracy.

关 键 词：抛光 材料去除率 BP神经网络 遗传算法 预测模型 

分 类 号：TG356.28[金属学及工艺—金属压力加工]