Research On Performance Degradation Prediction And Maintenance Strategy Of Turbofan Engines

In today’s competitive business world,one way to increase profitability of machinery equipment or a process plant is to reduce its operational and maintenance expenses while increasing productivity.Engine is one of the most expensive devices in aircraft and industrial applications,where reliability and availability are the two most desirable attributes.In the past several decades,trillions of dollars was invested globally in the o peration and maintenance of engines.However,due to their rising roles in the fast-growing industry,the market trend is still expected to be continued into the foreseeable future.Therefore,the development and implementation of a strong,efficient,flex ible maintenance strategy is an inevitable trend.At present,although the maintenance mode of aero-engine is moving forward from periodic overhaul and regular maintenance mode to timely repair,data processing and fault diagnosis in advance,intelligent algorithm and fault analysis of digital maintenance mode.However,even such digital maintenance is only a passive mode of m aintenance.With the development of related technology,intelligent maintenance will become a new trend after digital maintenance.As a new maintenance concept,intelligent maintenance is just at the beginning,and the research is just at the beginning.However,as the future direction of aero-engine health management,it will not change.At present,the research on this technology in C hina is still in the initial stage,and the relevant research is not very mature.This article focuses on the main technology of intelligent maintenance,followings are the relevant work:Firstly,to solve the problem of expensive data acquisition in relev ant research,the mechanism of the engine is modeled with modular modeling idea and the module that can characterize the degradation of each component is bui lt.The cause and influence of the degradation are analyzed in detail.Then,on the basis of the model,the modeling method of degradation propagation is studied and the sensor data generated by various components due to different reasons are simulated,which provides a data basis for the subsequent work.After that,the engine health status evaluation research was carried out,and the engine health parameter(HI),a parameter that can express the equipment health status,was established based on the sensor data,laying a foundation for the subsequent prediction work.Then,the short-term and long-term forecast of engine state is carried out for the short-term and long-term maintenance requirements.In the short-term trend prediction of engine,the HI is se lected as the prediction index to carry out the prediction.Through further analysis,it is found that this problem belongs to the trend prediction category of time series,so the long and short time memory network that can describe the timing dependent type is adopted to carry out the trend prediction.To solve the problem of inaccurate forecast trend due to insufficient data,a prediction model based on migration is proposed.When making long-term forecast engine,to improve the prediction accuracy,with the method of fingerprint diagram recognition of the machine under test degradation reasons,then select the remaining service life as a predictor,comprehensive analysis based on similarity and two commonly used prediction method based on regression,gaus sian process regression prediction model was proposed based on similarity,implements the determi nistic prediction and uncertainty of the residual lifetime prediction.Finally,a dynamic aperiodic active maintenance strategy is designed based on CBM on the basis of the above established prognostic model and the optimization index of operation cost minimization.Particle swarm optimization algorithm is used to design the dynamic aperiodic active maintenance methods to find a compromise in the maintenance cost and maintenance time of maintenance plan,and effectively avoid excessive maintenance or inadequate maintenance ensuring the reliability of the equipment operation and to maximize the availability of capital.