点击次数：

论文类型：期刊论文

发表时间：2015-08-01

发表刊物：RELIABILITY ENGINEERING & SYSTEM SAFETY

收录刊物：SCIE、EI、Scopus

卷号：140

页面范围：208-217

ISSN号：0951-8320

关键字：Predictive maintenance (PdM); Inertial Navigation System (INS); System reliability; GO Methodology; Remaining Useful Life (RUL)

摘要：The goal of this study is to propose a reliability centered predictive maintenance scheme for a complex structure Inertial Navigation System (INS) with several redundant components. GO Methodology is applied to build the INS reliability analysis model-GO chart. Components Remaining Useful Life (RUL) and system reliability are updated dynamically based on the combination of components lifetime distribution function, stress samples, and the system GO chart. Considering the redundant design in INS, maintenance time is based not only on components RUL, but also (and mainly) on the timing of when system reliability fails to meet the set threshold. The definition of components maintenance priority balances three factors: components importance to system, risk degree, and detection difficulty. Maintenance Priority Number (MPN) is introduced, which may provide quantitative maintenance priority results for all components. A maintenance unit time cost model is built based on components MPN, components RUL predictive model and maintenance intervals for the optimization of maintenance scope. The proposed scheme can be applied to serve as the reference for INS maintenance. Finally, three numerical examples prove the proposed predictive maintenance scheme is feasible and effective. (C) 2015 Elsevier Ltd. All rights reserved.