风电轴承套圈超声滚挤压表层物理力学性能预测模型  

Prediction model of physical and mechanical properties of ultrasonic roller extrusion surface of wind power bearing ring

作　　者：姚国林 徐红玉[1] 王晓强[1,2] 崔凤奎 苏涌翔[1,2] YAO Guo-lin;XU Hong-yu;WANG Xiao-qiang;CUI Feng-kui;SU Yong-xiang(School of Mechatronics Engineering,Henan University of Science and Technology,Luoyang 471003,China;Henan Province Collaborative Innovation Center for Advanced Manufacturing of Machinery and Equipment,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》2020年第7期109-116,共8页Journal of Plasticity Engineering

基　　金：国家自然科学基金资助项目(U1804145);国家重点研发计划“制造基础技术与关键部件”重点专项(2018YFB2000405)。

摘　　要：为了实现对风电轴承套圈超声滚挤压表层物理力学性能的合理控制,以静压力、进给速度、转速和振幅为加工参数进行超声滚挤压正交试验设计,采用遗传算法对BP神经网络算法进行改进,借助正交试验数据建立风电轴承套圈超声滚挤压表层物理力学性能预测模型,并验证预测模型的准确性,再应用BP神经网络预测模型分析各加工参数对风电轴承套圈表层物理力学性能的影响规律。结果表明:风电轴承套圈表层残余压应力试验值与预测值平均相对误差为3.9%;表面加工硬化程度平均相对误差3.31%,试验值与预测值相差较小,证明了BP神经网络预测模型的预测精度较高;预测结果表明随着静压力的增大,残余压应力和表面硬度呈增大趋势;随进给速度增大,残余压应力和表面硬度减小;转速增大使表面硬度增大,残余压应力减小;随着振幅的增加,残余压应力呈增大趋势,表面硬度呈先增大后减小的趋势。预测模型的预测规律与试验规律相一致,能够实现对风电轴承套圈超声滚挤压表层物理力学性能的预测。To achieve reasonable control of physical and mechanical properties of ultrasonic roller extrusion surface of wind power bearing ring,the static pressure,feed rate,rotation speed and amplitude were taken as the processing parameters to design ultrasonic roller extrusion orthogonal test,and the genetic algorithm(GA)was used to improve the BP neural network algorithm.The prediction model of the physical and mechanical properties of ultrasonic roller extrusion surface of wind power bearing ring was established based on the orthogonal test data,and the accuracy of the prediction model was verified.The influence laws of various processing parameters on the physical and mechanical properties of surface of wind power bearing ring were analyzed using the BP neural network prediction model.The results show that the average relative error between the experimental value and the predicted value of the surface residual compressive stress of wind power bearing ring is 3.9%;the average relative error of surface work hardening degree is 3.31%,the difference between the experimental value and the predicted value is small,which proves that the prediction accuracy of the BP neural network prediction model is high;the prediction results show that the residual compressive stress and surface hardness increase with the increase of static pressure;the residual compressive stress and surface hardness decrease with the increase of feed rate;the increase of the rotational speed makes the surface hardness increase and residual compressive stress decrease;with the increase of amplitude,the residual compressive stress tends to increase,and the surface hardness tends to increase first and then decrease.The prediction law of the prediction model is consistent with the experimental law,which can predict the physical and mechanical properties of ultrasonic roller extrusion surface of wind power bearing ring.

关 键 词：超声滚挤压 遗传算法 BP神经网络算法 物理力学性能 预测模型 

分 类 号：TH123[机械工程—机械设计及理论]