基于IGWO-SVM的超声辅助磨削Al_(2)O_(3)陶瓷表面粗糙度预测  

Surface roughness prediction of Al_(2)O_(3) ceramics machined by ultrasonic assisted grinding based on IGWO-SVM

作　　者：赵明利[1] 宋士杰 李博涵[1] ZHAO Mingli;SONG Shijie;LI Bohan(School of Mechanical and Power Engineering,Henan Polytechnic University,Jiaozuo454000,Henan,China)

机构地区：[1]河南理工大学大学机械与动力工程学院,河南焦作454000

出　　处：《河南理工大学学报（自然科学版）》2021年第6期117-121,共5页Journal of Henan Polytechnic University（Natural Science）

基　　金：国家自然科学基金资助项目(E51175153);河南理工大学博士基金资助项目(B2016-27)。

摘　　要：为预测超声辅助加工Al_(2)O_(3)陶瓷表面粗糙度,进行了超声辅助磨削Al_(2)O_(3)陶瓷试验。用改进的灰狼优化算法(IGWO)对支持向量机(SVM)进行参数优化,建立IGWO-SVM预测模型,并与PSO-SVM预测模型、CS-SVM预测模型、GWO-SVM预测模型进行比较。结果表明:IGWO-SVM预测模型预测值与试验值最大绝对误差值为0.411 9,最小绝对误差值为0.002 4,平均绝对误差值为0.145 6,平方相关系数为0.931 092,均方误差为0.000 399 8,相比PSO-SVM预测模型、CS-SVM预测模型、GWO-SVM预测模型,该模型具有更高的预测精度和可靠度,能够对超声辅助磨削Al_(2)O_(3)陶瓷表面粗糙度进行更精准的预测。In this work, the ultrasonic-assisted grinding of Al_(2)O_(3) ceramics was experimented to predict the surface roughness of ultrasonic-assisted grinding of Al_(2)O_(3) ceramics.The improved gray wolf optimization algorithm(IGWO)was used to optimize the parameters of the support vector machine(SVM),and the IGWO-SVM prediction model was established to predict the surface roughness.The IGWO-SVM model was compared with PSO-SVM prediction model, CS-SVM prediction model, and GWO-SVM prediction model.The results showed that the maximum absolute error between the predicted value of the IGWO-SVM prediction model and the experimental value was 0.411 9,the absolute minimum error was 0.002 4,the absolute average error was 0.145 6,the square correlation coefficient was 0.931 092,and the mean square error was 0.000 399 8.Compared with the PSO-SVM prediction model, the CS-SVM prediction model and GWO-SVM prediction model, the proposed model featured higher prediction accuracy and reliability, and surface roughness of ultrasonic-assisted grinding could be predicted more accurately.

关 键 词：超声加工 AL2O3陶瓷 表面粗糙度 支持向量机 

分 类 号：TG663[金属学及工艺—金属切削加工及机床] TG113.25