摘要
剩余寿命(Remaining Useful Lifetime,RUL)预测和预测维修决策是预测与健康管理(Prognostics and Health Management,PHM)研究的两大核心内容。RUL预测技术已取得丰硕的成果,预测维修决策研究仍不够深入。现有预测维修决策研究更多关注如何根据单次预测结果进行单次的维修决策。然而,预测是一个动态过程,相应的维修决策也是动态的,动态序贯的决策中决策的更新与停止问题是一个鲜有关注却值得研究的问题。针对以上问题,对状态可检测的系统,采用Bayesian更新和EM算法相结合的方法,实现了系统RUL和可靠性预测结果的动态更新;在此基础上制定序贯维修策略,依据自适应预测周期和RUL预测结果给出了决策停止的判断准则,在动态冗余的预测结果中选择出真正可执行的最优维修时刻;通过仿真实验验证了所提策略和方法的有效性和适用性。结果表明所提出的方法可有效降低系统的故障概率和运维成本,提升运维效率。
Remaining Useful Life(RUL)prediction has achieved many attentions,but the research on predictive maintenance decision is still not deep enough.The existing predictive maintenance decision research focuses on how to make a single decision based on the single prediction result.However,the prediction is a dynamic process,and the corresponding maintenance decision is also be dynamic.The renewal and stop of the decision in the dynamic sequential decision is a problem that is rarely concerned but worth studying.Aiming at the above problems,this paper combined Bayesian update and EM algorithm to achieve the dynamic update of RUL and predicted reliability of the system;On the basis of the dynamic adaptive RUL prediction,a sequential maintenance strategy is formulated.According to the adaptive detection and prediction cycle and the RUL prediction results,the judgment criteria for decision stopping are given,and the executable optimal maintenance time is decided from the prediction results of dynamic redundancy;The effectiveness and applicability of the proposed strategies and methods are verified through simulation experiments and case studies.The results indicated that the proposed method can effectively reduce the system failure probability and operation and maintenance cost through improve the operation and maintenance efficiency.
作者
张晓红
杨天祥
石冠男
秦彦凯
曾建潮
ZHANG Xiao-hong;YANG Tian-xiang;SHI Guan-nan;QIN Yan-kai;ZENG Jian-chao(Division of Industrial and System Engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China;College of Electrical andInformation Engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China;School of Mechanical Engineering,Taiyuan University of Science&Technology,Taiyuan 030024,China;Institute for Big Data and Visual Computing,North University of China,Taiyuan 030051,China)
出处
《太原科技大学学报》
2025年第5期411-419,共9页
Journal of Taiyuan University of Science and Technology
基金
国家自然科学基金项目(72071183)
山西省自然科学基金项目(202203021211194)
山西省回国留学人员科研资助项目(2022-161)。
关键词
动态RUL预测
动态序贯决策
预测维修决策
决策停止
dynamic RUL prediction
dynamic sequential maintenance strategy
predictive maintenance decision
decision stopping
