摘要
对影响刀具寿命的因素进行分析,确定了影响刀具寿命的主要影响因素,建立基于BP-RBF神经网络的刀具寿命预测模型;对刀具试验寿命数据样本进行统计,采用最小二乘法对刀具寿命预测数学模型进行非线性拟合,建立试验刀具寿命模型。通过十折交叉验证方法对BP-RBF神经网络模型和传统BP神经网络模型进行试验仿真,结合刀具寿命数据样本对刀具寿命模型进行验证。通过与传统BP神经网络模型和刀具寿命预测模型对比可得:BP-RBF神经网络具备更高的预测精度,该预测模型在刀具寿命预测上具备有效性。
Based on the traditional tool life prediction formula,the main influencing factors of tool life are determined through the analysis of the influencing factors of tool life,and a tool life prediction model based on BP-RBF neural network is established;the tool life data samples of experiments are counted,the least square method is used to nonlinearly fit the mathematical model for establishing the experimental tool life model.The BP-RBF neural network model and the traditional BP neural network model are tested and simulated through a 10-fold cross-validation method,and the tool life data samples are combined to verify the established tool life model.By comparing with the traditional BP neural network model and tool life prediction model,the BP-RBF neural network has higher prediction accuracy,and the prediction model is effective in tool life prediction.
作者
方喜峰
张杰
程德俊
张胜文
Fang Xifeng;Zhang Jie;Cheng Dejun;Zhang Shengwen(School of Mechanical Engineering,Jiangsu University of Science and Technology,Zhengang,Jiangsu 212003,China;不详)
出处
《工具技术》
2020年第12期69-69,70-73,共5页
Tool Engineering
基金
国防基础科研基金项目(A0720133010)
江苏省先进制造技术重点实验室开放基金资助(HGAMTL-1905)
镇江市重点研发计划项目(GY2019003)。
