现代制造工程 ›› 2024, Vol. 523 ›› Issue (4): 140-145.doi: 10.16731/j.cnki.1671-3133.2024.04.019

• 设备设计/诊断维修/再制造 • 上一篇    下一篇

基于BP神经网络的谐波减速器柔轮疲劳寿命预测研究*

成元彬1, 袁文平1, 张涛2, 刘志峰3   

  1. 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
  • 收稿日期:2023-09-04 出版日期:2024-04-18 发布日期:2024-05-31
  • 作者简介:成元彬,高级工程师,研究方向为机器人及智能装备。E-mail:cyb369@qq.com
  • 基金资助:
    *北京市科学技术研究院装备改革与发展专项项目(0820230195KF001-01)

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

CHENG Yuanbin1, YUAN Wenping1, ZHANG Tao2, LIU Zhifeng3   

  1. 1 Institute of lntelligent Equipment,Beijing Academy of Science and Technology, Beijng 100061,China;
    2 Department of Materials and Manufacturing,Beijing University of Technology, Beijng 100124,China;
    3 School of Mechanical and Aerospace Engineering, Jilin University,Changchun 130025,China
  • Received:2023-09-04 Online:2024-04-18 Published:2024-05-31

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。

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

Abstract: 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.

Key words: flexspline, stress analysis, fatigue life prediction, finite element, Back Propagation (BP) neural network

中图分类号: 

版权所有 © 《现代制造工程》编辑部 
地址:北京市东城区东四块玉南街28号 邮编:100061 电话:010-67126028 电子信箱:2645173083@qq.com
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn