超声滚挤压轴承套圈表面粗糙度数学模型对比分析  

Comparative analysis of mathematical model for surface roughness of ultrasonic rolling extrusion bearing rings

作　　者：崔凤奎[1,2] 苏涌翔 荣莎莎 姚国林 CUI Feng-kui1,2, SU Yong-xiang1,2, RONG Sha-sha1,2, YAO Guo-lin1,2(1. School of Mechatronics Engineering, Henan University of Science and Technology, Luoyang 471003, China; 2. Collaborative Innovation Center of Machinery, Equipment Advanced Manufacturing of Henan Province, Luoyang 471003, China)

机构地区：[1]河南科技大学机电工程学院,河南洛阳471003 [2]机械装备先进制造河南省协同创新中心,河南洛阳471003

出　　处：《塑性工程学报》2018年第5期199-204,共6页Journal of Plasticity Engineering

基　　金：国家自然科学基金资助项目(51475146; 51475366)

摘　　要：为了提高超声滚挤压轴承套圈的表面性能,以轴承套圈为研究对象,通过超声滚挤压试验,对试验结果进行数理统计分析,研究加工参数对超声滚挤压轴承套圈表面粗糙度的影响;再对超声滚挤压试验进行正交试验设计,将超声滚挤压轴承套圈表面粗糙度与各加工参数相互匹配,建立响应曲面和BP神经网络模型,两个模型试验结果与预测结果的对比表明所建立的轴承套圈表面粗糙度BP神经网络模型的相对误差控制在4. 5%左右,最大误差不超过5. 06%,预测结果具有更高的可信度,且优于响应曲面模型预测结果,可以进行不同超声滚挤压参数的轴承套圈表面粗糙度的预测。In order to improve the surface properties of ultrasonic rolling extrusion bearing rings,taking bearing ring as the object,the influences of processing parameters on the surface roughness were studied by using experimental and statistical methods. Based on the orthogonal design of ultrasonic roller extrusion test,the surface roughness of the bearing ring and the corresponding processing parameters were mutually matched,and then the response surface model and the BP neural network model were established. The comparison of predicted results with the two models shows that the established BP neutral network model for surface roughness presents a good agreement with the experiment and is better than the surface response model in surface roughness prediction. The relative error of BP neural network model is about 4. 5%,and the maximum error is less than about 5. 06%,which can be used for surface roughness prediction of ultrasonic roll extrusion bearing rings with different parameters.

关 键 词：轴承套圈 超声滚挤压 表面粗糙度 响应曲面 BP神经网络 预测模型 

分 类 号：TG115[金属学及工艺—物理冶金]