考虑相关性和任务约束的复杂系统动态维修优化

Modern industrial systems and products are more advanced in technology, more complex in composition, and more prominent in effect of civil and defense fields. Maintenance is an important means to ensure the safe and reliable operation of the system. Many studies have been presented on the multi-component maintenance; however, there are some problems need to be further explored. Thus, this project will study the following aspects step by step. (1)The existing multi-component maintenance optimization rarely touches on the topic of stochastic dependency and the operation station. So a condition-based maintenance (CBM) optimization for the two component system with degradation rate interactions is presented firstly. After that, a rolling horizon method is used to study the dynamic preventive maintenance (PM) optimization for the multi-component systems in consideration of stochastic dependencies and the operation station.(2)There is a closing link between maintenance activities and spare parts management. And the spare part is an important factor affecting maintenance. For both multi-component systems with a known number of identical and non-identical components, the joint optimization of spare parts ordering with CBM and PM are studied, respectively. (3)Aiming at the characteristics of the equipment, the single mission oriented dynamic maintenance optimization based on the above two studies. The research will provide new theory for the multi-component system maintenance optimization and help the enterprise make decision-making more scientific, improving the equipment reliability and economic benefits.

现代工业系统以及产品的技术更加先进，构成更加复杂，在民用以及国防领域的作用也更加突出，维修是确保系统安全可靠运行的重要手段。多部件系统维修优化研究目前已取得一定成果，但仍有一些问题需要解决。为此，本项目针对多部件系统从三方面开展研究：（1）现有的多部件系统维修优化较少考虑随机相关和运行状态信息，为此,提出考虑退化状态相关的两部件系统视情维修优化，而后利用移动时间窗理论，研究考虑运行状态和随机相关的多部件系统动态预防性维修优化；（2）备件与维修密切相关，是影响维修的重要因素，为此，研究建立相同多部件系统视情维修与备件订购联合优化，并进一步考虑不同多部件系统预防性维修与备件供应联合优化（3）针对任务型装备特点，结合（1）和（2）的研究，建立面向单阶段任务的多部件系统动态维修优化。该项目将进一步丰富多部件系统维修决策优化理论，帮助企业进行科学决策以提高设备的可靠性和经济效益。

现代工业系统以及军用装备的技术越来越先进，大部分为多部件组成的复杂系统，在民用以及国防领域的作用也更加突出，维修是确保系统安全可靠运行的重要手段。多部件系统的部件间由于受内外部环境影响往往存在相关性，此外，系统在执行任务过程中可能会受到多种因素影响。为此，本项目针对多部件系统开展了以下研究：（1）基于Copula函数描述多部件系统的退化相关，在此基础上建立了多部件系统动态视情维修优化模型；（2）备件与维修密切相关，是影响维修的重要因素，为此，在多部件系统视情维修的基础上建立了维修与备件订购联合优化方法；（3）针对任务型装备特点，初步建立面向单阶段任务的多部件系统维修优化；（4）拓展研究了故障预测与健康管理相关的智能故障诊断方法以及备件供应保障优化方法，有助于项目理论向实践的转化。项目研究成果进一步丰富了多部件系统维修优化理论，有助于提高系统维修保障的科学性。