系统工程与电子技术 ›› 2021, Vol. 43 ›› Issue (9): 2688-2696.doi: 10.12305/j.issn.1001-506X.2021.09.40

• 可靠性 • 上一篇    

基于改进相似性的装备部件剩余寿命预测及经济性储备策略

陈云翔, 饶益*, 蔡忠义, 王泽洲   

  1. 空军工程大学装备管理与无人机工程学院, 陕西 西安 710051
  • 收稿日期:2021-12-29 出版日期:2021-08-20 发布日期:2021-08-26
  • 通讯作者: 饶益
  • 作者简介:陈云翔 (1962—), 男, 教授, 博士, 主要研究方向为装备管理与决策、装备系统工程|饶益 (1997—), 男, 硕士研究生, 主要研究方向为装备发展与管理决策|蔡忠义 (1988—), 男, 讲师, 博士, 主要研究方向为装备可靠性与系统工程|王泽洲 (1992—), 男, 博士研究生, 主要研究方向为装备可靠性评估、剩余寿命预测
  • 基金资助:
    国家自然科学基金(71901216)

Remaining useful lifetime prediction and economic reserve strategy of equipment components based on improved similarity

Yunxiang CHEN, Yi RAO*, Zhongyi CAI, Zezhou WANG   

  1. Equipment Management and Unmanned Aerial Vehicle Engineering College, Air Force Engineering University, Xi'an 710051, China
  • Received:2021-12-29 Online:2021-08-20 Published:2021-08-26
  • Contact: Yi RAO

摘要:

针对传统基于相似性的剩余寿命(remaining useful lifetime, RUL)预测方法未考虑运行条件差异, 从而影响预测准确性及部件储备策略科学性的问题, 提出一种基于改进相似性的装备部件RUL预测及经济性储备策略。基于提出的改进相似性方法, 区分装备部件的运行条件类别, 通过各类别内服役部件和参考部件的性能状态相似性, 预测服役部件的RUL; 基于RUL预测结果, 以装备部件维修储备总费用最低为目标, 以资源利用率为约束, 建立经济性储备策略决策模型; 采用差分进化算法对模型寻优求解, 得到最优装备部件储备策略。实例分析表明, 所提方法能够有效提升RUL预测的准确性和部件储备策略的科学性, 具备工程应用价值。

关键词: 剩余寿命预测, 储备策略, 改进相似性, 遗传算法-反向传播, 差分进化算法

Abstract:

Aiming at the problem of the traditional similarity-based remaining useful lifetime (RUL) prediction method does not consider the difference of operating conditions, which affects the accuracy of prediction and the scientific of components reserve strategy, an improved similarity-based RUL prediction and economic reserve strategy of equipment components is proposed. Based on the proposed improved similarity method, distinguishing the operating condition categories of equipment components, the RUL of the service components is predicted by the performance state similarity between the service components and the reference components in each category. Based on RUL prediction results, the economic reserve strategy decision model is established by taking the lowest total cost of equipment components maintenance and reserve as the objective and the resource utilization as the constraint. The differential evolution algorithm is used to optimize the model to obtain the optimal equipment components reserve strategy. The results of numerical example analysis show that the proposed method can effectively improve the accuracy of the RUL prediction and the scientific of components reserve strategy, which has engineering application value.

Key words: remaining useful lifetime (RUL) prediction, reserve strategy, improved similarity, GA-BP (genetice algorithm-back propagation), differential evolution algorithm

中图分类号: