Fault diagnosis and prognosis based on physical knowledge and reliability data: Application to MOS Field-Effect Transistor

摘要

The reliability data are very useful in maintenance operations since they are used to calculate the Mean Time To Failure. However, they are rarely used for the online calculation of the Remaining Useful Life (RUL). This is because the features measured in the reliability tests are not always measurable online. In this paper, the proposed solution is usage of physical knowledge of the components to build models linking these features to the variables measurable online. Then, the physical models are used to generate health indices whose evolution can be estimated and predicted online, and the reliability data used for initializing the trend models of the health indices. To guarantee the robustness of the remaining useful life estimation to changes in Condition Monitoring, this paper proposes to update online the drift parameter of Wiener process. The latter is used to model the trend of the health indices. The results obtained by the most used updating methods in the literature are analyzed and compared in order to highlight the influence of the model updating on prognosis performance. Experimental results, obtained by an application to estimate the RUL of MOS Field-Effect Transistor, show the effectiveness of the proposed method.