首页|基于GA-SVM的超声切削力预测模型研究

基于GA-SVM的超声切削力预测模型研究

扫码查看
Nomex蜂窝材料是难加工材料,选择合适的加工方法和加工参数对于控制切削力和提高加工质量非常重要,开展了超声辅助切削Nomex蜂窝材料的三维有限元仿真,对前期的超声切削Nomex蜂窝材料加工试验进行分析,针对影响切削力的加工参数设计正交试验,构建基于支持向量机(SVM)的超声切割切削力预测模型,提出一种遗传算法(GA)优化参数方法,设定该模型的最终真实值与预测值的相对误差在 0~3.55%范围内且R2=0.996 6,选取 4 组不同于正交试验表的加工参数进行验证.结果表明,该方法有效可行,对选择合适的加工参数、减小切削力具有积极意义.
Research on Ultrasonic Cutting Force Prediction Model Based on GA-SVM
Nomex honeycomb material is a difficult-to-machine material.Cutting force is one of the indicators for evaluating cutting processability.Choosing appropriate processing methods and processing parameters is very important to control the cutting force and improve processing quality.Through the three-dimensional finite element simulation of ultrasonic cutting of Nomex honeycomb materials and the analysis of previous ultrasonic cutting experiments on Nomex honeycomb materials,an orthogonal experiment was designed for the processing parameters that affect the cutting force,a cutting force prediction model for ultrasonic cutting based on support vector machine(SVM)was constructed,and a genetic algorithm(GA)optimization parameter method was proposed.The relative error between the final true value and the predicted value of the model is in the range of 0~3.55%,R2=0.996 6.By selecting four groups of machining parameters different from the orthogonal test table for verification,the results show that this method is effective and feasible,and has positive significance for selecting appropriate machining parameters and reducing cutting forces.

GA-SVMultrasonic cuttingfinite element simulationprediction model

李恒锋、何宇、邹平、王安斌

展开 >

上海工程技术大学城市轨道交通学院,上海 201620

东北大学机械工程与自动化学院,辽宁沈阳 110819

GA-SVM 超声切削 有限元仿真 预测模型

2024

电加工与模具
苏州电加工机床研究所 中国机械工程学会特种加工分会

电加工与模具

CSTPCD北大核心
影响因子:0.285
ISSN:1009-279X
年,卷(期):2024.(5)
  • 5