Aeroengine Remaining Life Prediction Algorithm Based on Improved Differential Time Domain Features and LSTM

In order to ensure the continuous airworthiness of the engine, airlines must carry out maintenance, repair and overhaul of the engine. This paper studies the prediction of the residual life of the aeroengine based on the improved differential time domain feature and LSTM, and analyzes the prediction framework, model and related algorithms of the residual life of the aeroengine based on the improved differential time domain feature and LSTM. This paper builds an engine life prediction algorithm DTF-LSTM based on improved differential time-domain features (DTF) and LSTM network. The network directly enhances the inheritance of historical output information by adding linear connections between adjacent output layers. The abstract local features extracted by LSTM are used as the input of the regression to predict the remaining life of the aero-engine. The predicted value of DTF-LSTM is close to the real value, and fitting the predicted value can obtain the residual service life curve of the aero-engine, which can accurately judge the degree of bearing degradation.