具有隐藏故障的多状态竞争失效系统检测策略研究

复杂工业系统的故障中有40%属于隐藏故障，若隐藏故障得不到及时消除，则可能导致巨大的经济损失。本文针对具有隐藏故障的多状态竞争失效系统，在考虑不完全检测的基础上，对其进行了维修建模。首先，基于制定的视情维修策略，分析了系统的更新情形；然后，基于所推导出的各种系统更新情形的发生概率，建立了以期望成本率为目标函数，以检测间隔时间为决策变量的维修模型。最后，以电容器组为例对所构建维修模型的适用性和有效性进行了验证，研究结果表明，检测质量对检测策略的制定有重要影响，因此研究具有隐藏故障的多状态竞争失效系统的检测策略时，有必要考虑检测误差。本文提出的维修建模方法不仅能在一定程度上丰富多状态竞争失效系统的维修策略理论与方法，还能为工程实际中复杂工业系统的维修工作提供理论基础与决策依据。

Inspection Strategy for a Multi-state Competing Failure System with Hidden Failures

40% of all failure modes of complex industrial systems fall into the category of hidden failures. It is likely to cause huge economic losses if these hidden failures are not eliminated in time. In this paper, the maintenance strategy problem of a multi-state competing failure system with hidden failures is studied by considering imperfect inspection. Firstly, all the renewal scenarios of the system are analyzed according to the given condition-based maintenance strategy. Then, the occurrence possibility of each renewal scenario is derived, based on which the maintenance model is established. The constructed maintenance model takes the expected cost rate as the objective and the inspection interval as the optimization variable. Finally, a capacitor bank is taken as an example to verify the applicability and effectiveness of the proposed maintenance model. The results show that inspection quality has an important influence on the decision-making of the inspection strategy. Therefore, inspection errors should be considered when constructing maintenance model. The modeling method proposed in this paper can not only enrich the maintenance strategy theory and method of multi-state competing failure systems to a certain extent, but also provide theoretical basis and decision-making basis for the maintenance of complex industrial systems in engineering practice.