Prognostics and health management for electromechanical system: A review

Abstract

As a transmission component, gears take on a great significance for the Electromechanical system of aviation equipment and has long aroused the widespread attention of researchers. Fault diagnosis and remaining useful life (RUL) prediction during the gear operation is critical to prognostics and health management (PHM) of gear transmission systems. In this paper, the focus is placed on gear PHM methods. This paper attempts to review the existing methods and summarize them into four types (including physical model-based, knowledge modelbased, data-driven model-based, as well as hybrid model-based methods). Based on a wide variety of methods, the principle and the application situation are indicated. In particular, the data-driven model-based methods consist of stochastic algorithms, statistical algorithms, as well as the artificial intelligence (AI) method. The fault diagnosis, performance degradation and RUL prediction of various methods are primarily summarized. Furthermore, the advantages and disadvantages of various methods are assessed, and the prospect of the Digital Twin (DT) is forecasted to boost the applications of PHM.