Remaining useful life estimation of aeroengine based on MPFAM-FCN

As the core component of an aircraft, an aero-engine provides power to the aircraft and its reliability is the key to ensure the safety of the aircraft. The remaining useful life prediction is of great significance to improve the availability of aero engines and reduce their maintenance costs. To achieve end-to-end RUL prediction and improve the prediction accuracy, this paper proposes an aero-engine remaining life prediction method based on the MPFAM-FCN model, and experiments are conducted on the dataset of C-MAPSS to verify the effectiveness of the proposed method. First, the C-MAPSS dataset is preprocessed; then the training set is input to the MPFAM-FCN prediction model for training; finally, the validation comparison is carried out based on the C-MAPSS dataset. The results show that the model can accurately predict the remaining service life of an aero-engine, and its prediction is also better than some existing advanced algorithms.

I. Introduction

With the development of science and technology, mechanical equipment is becoming more and more integrated, and traditional repair and maintenance strategies can no longer meet the demand for efficient and accurate maintenance. Machinery and equipment operating under relatively harsh working conditions also leads to a certain degree of wear and degradation of its internal parts. For some important machinery and equipment such as aircraft engines, rolling bearings, etc., failure may cause serious economic losses or even casualties. Remaining useful life (RUL) prediction as an important part of prognostics and health management (PHM) technology has received much attention from experts and scholars in recent years, among which, aero-engine is one of the key objects of concern in daily aviation maintenance operations. Achieving accurate RUL prediction for aero engines is of great importance for companies to make reasonable maintenance plans and reduce maintenance costs. At present, there are two main types of methods to predict the RUL of aero engines: physical model-based prediction methods and data-driven prediction methods [1].