论文标题
系统级预测维护：研究文献和差距分析的审查
System-Level Predictive Maintenance: Review of Research Literature and Gap Analysis
论文作者
论文摘要
本文从系统的角度回顾了预测维护领域的当前文献。我们将目前适用于简单组件的状况估计和故障风险预测的现有功能与解决复杂资产的相同任务所需的功能。系统级分析面对更复杂的潜在降解状态，它必须在每个组件级别上全面考虑主动维护程序，并考虑在不同的维护措施之间进行耦合，同时反映了系统故障的增加的货币和安全成本。结果，有效预测风险并告知有关各个组件的维护决策的方法不会容易扩展以提供可靠的子系统或系统级别的见解。需要采用一种新型的整体建模方法来融合可用的结构和物理知识，并自然处理积极进行和维护资产的复杂性。
This paper reviews current literature in the field of predictive maintenance from the system point of view. We differentiate the existing capabilities of condition estimation and failure risk forecasting as currently applied to simple components, from the capabilities needed to solve the same tasks for complex assets. System-level analysis faces more complex latent degradation states, it has to comprehensively account for active maintenance programs at each component level and consider coupling between different maintenance actions, while reflecting increased monetary and safety costs for system failures. As a result, methods that are effective for forecasting risk and informing maintenance decisions regarding individual components do not readily scale to provide reliable sub-system or system level insights. A novel holistic modeling approach is needed to incorporate available structural and physical knowledge and naturally handle the complexities of actively fielded and maintained assets.
