Research On Methods Of Performance Reliability Assessments And Life On Wing Prediction For Civil Aeroengine

By way of airplane heart, the health of aero engines has important signification for the safe of airplane and the decrease of operation cost. The reliability assessments for civil aero engines are significant in the process of total aero engine's design, manufacturing, use, maintenance and so on. The accurate reliability assessments can effectively direct the aero engines use and maintenance assignment, and establish the base for improving the airplane's safety and economy operation. Traditional reliability theory makes the reliability assessments depending on failure data. However, it's difficult to obtain enough failure data in allowable time. So under the condition of few or no failures, the reliability assessments for civil aero engines have become the key questions of aviation field. However, at the 20 century end, the performance reliability analysis theory and methods based on performance degradation data provide an effective way for solve the question. With the background of the situation, using the collected performance data in the process of aero engines use and maintenance, based on the performance reliability analysis theory, this paper puts emphasis on the researches about the theory and technique of aero engine's performance reliability assessments. The main research contents and conclusions include:(1) The reliability assessment method for civil aero engine fleet based on single performance parameter is proposed. Firstly, collecting the performance data in the aero engines use process, through influence factors analysis, based on stochastic process theory, the performance degradation model is established. At the same time, through analyzing the relation between the performance degradation process and life distribution, using Kolmogorov-forward-equation and the Brownian motion property, the first-arrival time distribution is established. Consequently, the aero engines reliability could be real time evaluated. Very high evaluation precision is obtained. The method shows superiority at the reliability analysis under few failures condition.(2) Traditional life prediction methods are based on a mass of life information. However, for high reliability aero engines, it's very difficult to collect enough life data. For aero engines,it is possible to obtain degradation measurements over time and these measurements may contain useful information about aero engine reliability. So, in this paper, based on the performance reliability analysis theory, using K-S test to obtain the prior distribution of models, the aero engine performance degradation models are respectively established by Bayesian updating methods and time series analysis methods. Thereby, the single life prediction information under failure limit is obtained. Finally, the life on wing combined prediction model is presented by using IA-PSO method. And through comparing the prediction results, the combined prediction precision is distinctly improved. This method could predict the life of products being just now using but no life data, and has comprehensive applicability.(3) The aero engine is a very complicated electromechanical system. Its health shows on multi-performance parameters, and it have relativity among these parameters. For completely evaluating engines reliability, the aero engine real-time performance reliability evaluation method and life control method are presented in this paper. First, through analyzing the performance degradation process of the engine, the time-varing performance degradation model based on state space method was built, and performance trends were predicted by using Kalman filtering. Then, considering the correlation of multiple performance parameters, the joint probability distribution based on multi-performance parameters is established, and consequently engines performance reliability could be evaluated by real time. Comparing with traditional reliability evaluation methods, the above method can evaluate the single engine performance reliability by real time. Airline Company may control engine life according to engine different performance reliability levels.(4) Due to aero engines having multi-failure modes, its failure is the competition result of different failures. So,in the paper, the aero engine reliability evaluation model based on multi-failure modes is proposed. Fist, depending on (3),the aero engine degradation failure model based on multi-performance parameters is established. Then, for typical failures, because of the correlation of these failures mode and that proportion hazard model could commendably incarnate the influence which performance parameters have for the other typical failures, the typical failures models based on proportion hazard model are established. Finally, according to reliability analysis theory, the aero engine reliability evaluation model based on multi-failure modes is presented. This method can provide which failure mode may be easier to arouse engines failure and the probability of the failure mode. For airline company, they can predominate the engine health in time, institute the reasonable maintenance plan and flexible manage engine fleet.(5) Combining Airlines practical works, an integrated aeroengine maintenance management system is designed. This system can implement engine condition monitoring, LOW control, maintenance decision-making and optimal engine schedule and is being applied to the aeroengine engineering management in airlines effectively.