基于BP神经网络的谐波减速器柔轮疲劳寿命预测研究  

Research on fatigue life prediction of flexspline of harmonic reducer based on BP neural network

作　　者：成元彬 袁文平 张涛[2] 刘志峰 CHENG Yuanbin;YUAN Wenping;ZHANG Tao;LIU Zhifeng(Institute of lntelligent Equipment,Beijing Academy of Science and Technology,Beijng 100061,China;Department of Materials and Manufacturing,Beijing University of Technology,Beijng 100124,China;School of Mechanical and Aerospace Engineering,Jilin University,Changchun 130025,China)

机构地区：[1]北京市科学技术研究院智能装备研究所,北京100061 [2]北京工业大学材料与制造学部,北京100124 [3]吉林大学机械与航空航天工程学院,长春130025

出　　处：《现代制造工程》2024年第4期140-145,共6页Modern Manufacturing Engineering

基　　金：北京市科学技术研究院装备改革与发展专项项目(0820230195KF001-01)。

摘　　要：柔轮是谐波减速器的易损零件,在波发生器的高转速带动下转动,其疲劳寿命一直是备受关注的研究重点。以某型号杯型谐波减速器柔轮为研究对象,建立有限元仿真模型,得到柔轮最大应力与筒长、筒体壁厚和不同过渡圆角半径等参数之间的关系。根据柔轮S-N曲线,计算得到柔轮疲劳寿命,利用反向传播(Back Propagation,BP)神经网络实现了柔轮疲劳寿命的预测。The flexspline is a vulnerable part of the harmonic reducer.Driven by the high rotation speed of the wave generator,its fatigue life has always been a research focus of concern.Takes flexible wheel of a cup-type harmonic reducer as the research object,establishes a finite element simulation model,the relationship between its maximum stress and the parameters such as the length of the cylinder,the thickness of the cup wall and the radius of different transition fillets were obtained.Based on the S-N curve of the flexspline,the fatigue life of the flexspline was calculated,and the prediction of fatigue life was achieved using BP neural network.

关 键 词：柔轮 应力分析 疲劳寿命预测 有限元分析 反向传播神经网络 

分 类 号：TH132.43[机械工程—机械制造及自动化]