China Mechanical Engineering

Previous Articles     Next Articles

Prediction Model of Surface Roughness of 8418 Steel by EDM Based on SVM

YU Jianwu;HU Qifeng;WEN Cheng;LIU Bo;SHEN Xiang   

  1. College of Mechanical & Vehicle Engineering, Hunan University,Changsha,410082
  • Online:2018-04-10 Published:2018-04-03
  • Supported by:
    National Natural Science Foundation of China(No. 51275165)
    Hunan Provincial Natural Science Foundation of China(No. 2015JJ2026)

基于支持向量机的电火花加工8418钢表面粗糙度预测模型

余剑武;胡其丰;文丞;柳波;沈湘   

  1. 湖南大学机械与运载工程学院,长沙,410082
  • 基金资助:
    国家自然科学基金资助项目(51275165);
    湖南省自然科学基金资助项目(2015JJ2026)
    National Natural Science Foundation of China(No. 51275165)
    Hunan Provincial Natural Science Foundation of China(No. 2015JJ2026)

Abstract: Because of the complexity of EDM processes, it took a great deal of time and high experimantal costs to study the effects of various discharge parameters and non-electric parameters on Ra of the workpieces. Based on SVM, a model was proposed to predict the surface roughness of workpieces after EDM, and genetic algorithm was used to optimize the parameters of the model to predict the surface roughness machined by different electrical discharge parameters. An EDM experiment was applied on 8418 steel to verify the accuracy of prediction model of surface roughness by comparing the predicted values with the experimental ones. Finally the errors analysis were carried on, and the maximun error of model is 2.27%.

Key words: electrical discharge machining(EDM), surface roughness, prediction model, support vector machine(SVM), 8418 steel

摘要: 由于电火花加工过程的复杂性,单纯通过电火花加工实验方法研究各种放电参数及非电参数对工件表面粗糙度Ra的影响不但耗费大量时间,而且实验成本较高,为此基于支持向量机提出了一种适用于电火花加工表面粗糙度预测的模型。利用遗传算法对该模型中的各参数进行优化,预测不同电火花加工参数组合下的表面粗糙度;以电火花加工8418模具钢为例,将预测值与实验值进行对比,并且通过实验验证了电火花加工8418钢表面粗糙度预测模型参数的准确性;最后进行了误差分析,模型的最大误差值为2.27%。

关键词: 电火花加工, 表面粗糙度, 预测模型, 支持向量机, 8418钢

CLC Number: