Design Of Performance Evaluation System For Aero-engine Based On Information Granular

Aero engine is the core power unit of aircraft.Prognostics and health management of aero engine are very important for improving safety and reducing flight cost.The prediction of remaining useful life is a hot topic in health management of aero engine in recent years.Therefore,based on the information granular theory and technique,this paper studies the prediction method of remaining useful life of aero engines.The main work is as follows.Firstly,the multi-sensor monitoring data of engine is divided into several intervals based on information granular theory.The degradation interval of the engine is found and the maximum remaining useful life of each engine is set dynamically according to it.The piece-wise degradation model of the engine performance is constructed based on the performance degradation feature.The data set of engine is used to verify the effectiveness of the proposed method.The experiment shows that dynamically setting the maximum remaining useful life according to the degradation interval can obtain more accurate result.Secondly,an interval forecasting model based on the multi-layer perceptron for aero engine is proposed.The interval-valued remaining useful life data set is constructed by using the center and radius of the degradation interval.The interval-valued remaining useful life is predicted by the model of upper bound and lower bound trained respectively.The experiment of prediction of interval-valued remaining useful life is conducted using the data set of engine.The result shows that the proposed model basically covers the actual remaining useful life.Thirdly,according to the requirements of the aero engine design and maintenance,the aero engine data management platform is built.There are data management-maintenance modules and engine performance evaluation modules,which can implement the following functions: import and export of aero engine data,query and update of logbook,visual preview of performance parameters and so on.At the same time,the performance of aero-engine is evaluated by combining the data analysis functions such as remaining useful life prediction.The evaluation results will be exported to the report,which will provide important reference for engine design and maintenance.