计算机集成制造系统 ›› 2021, Vol. 27 ›› Issue (4): 1052-1061.DOI: 10.13196/j.cims.2021.04.009

• 当期目次 • 上一篇    下一篇

基于神经网络和再制造因子的再制造机床可靠性分配方法

杜彦斌1,2,吴国奥2,3,许磊1   

  1. 1.重庆工商大学制造装备机构设计与控制重庆市重点实验室
    2.重庆工商大学国家智能制造服务国际科技合作基地
    3.重庆工商大学管理科学与工程学院
  • 出版日期:2021-04-30 发布日期:2021-04-30
  • 基金资助:
    国家自然科学基金资助项目(51775071);重庆工商大学研究生创新型科研资助项目(yjscxx2020-094-60);重庆市教委科学技术研究计划重点项目(KJZD-K202000801);重庆工商大学重点科研平台资助项目(KFJJ2019073)。

Reliability allocation method for remanufactured machine tools based on neural network and remanufacturing factor

  • Online:2021-04-30 Published:2021-04-30
  • Supported by:
    Project supported by the National Natural Science Foundation,China(No.51775071),the Innovation Foundation for Graduate Students of CTBU,China(No.yjscxx2020-094-60),the Science and Technology Research Program of Chongqing Municipal Education Commission,China(No.KJZD-K202000801),and the Key Research Platform of CTBU,China(No.KFJJ2019073).

摘要: 为了保障和提高再制造机床的可靠性,提出一种基于神经网络和再制造因子的再制造机床可靠性分配方法。该分配方法根据可靠性逐级分配的思想,建立再制造机床故障树模型,将再制造机床的故障划分为系统级、子系统级和零部件级;训练三层前馈神经网络,将整机可靠性目标从系统级分配到子系统级;基于多层模糊评价确定再制造因子,提出再制造因子与概率重要度结合的可靠性分配方法,将可靠性从子系统级分配到零部件级;以数控滚齿机再制造为工程算例,将可靠性目标分配到再制造滚齿机各个零部件。结果表明,该方法可以保证再制造机床达到可靠性目标并实现可靠性提升。

关键词: 再制造机床, 可靠性分配, 故障树模型, 神经网络, 再制造因子

Abstract: To guarantee and improve the reliability of remanufactured machine tools,a reliability allocation method based on neural network and remanufacturing factor was proposed.According to the principle of reliability allocation level by level,the fault tree model of remanufactured machine tools was established,and the faults of remanufactured machine tools were divided into system-level,subsystem-level and part-level.The three-layer feedforward neural network was trained to allocate system reliability to subsystem-level.Remanufacturing factors were determined based on multi-hierarchy fuzzy assessment,and a reliability allocation method combining remanufacturing factors and probability importance was proposed to allocate reliability from subsystem-level to part-level.Taking the remanufacturing of NC gearhobbing machine as an engineering example,the reliability target was allocated to each component of the remanufactured gearhobbing machine.The result showed that the method could ensure the reliability of the remanufactured machine tools and improve their reliability.

Key words: remanufactured machine tools, reliability allocation, fault tree model, neural network, remanufacturin factors

中图分类号: