A Method for Remaining Life Prediction of Aviation Turbofan Engine Based on Multi-source Data Fusion

Abstract

The invention discloses a method for predicting the remaining life of a multi-source data driven aero-turbofan engine. According to the method, the remaining life of the engine is predicted by completely utilizing monitoring data acquired by an engine sensor. The method comprises four steps of estimating multi-source monitoring data fusion and failure threshold values; modeling the degradation process of the engine and estimating parameters; describing the remaining life of the engine; and predicting the remaining life. Compared with the prior art, the method has the following advantages: multi-source monitoring data is fused on the basis of common main component analysis and Euclidean distance to extract the health index and the failure threshold value representing the operating state of the engine and solve the problem that monitoring data information in the traditional prediction method is utilized incompletely; the Wiener process with nonlinear drift is established to represent the aero-engine degradation process with uncertain degradation rate, and real-time on-line life prediction is conducted on the basis of the aero-engine remaining life distribution; and data support is provided for on-condition maintenance, the time on wing of the engine is increased, major accidents are avoided, and engineering application value is achieved.

Description

technical field [0001] The invention relates to the field of forecasting the remaining life of an aviation turbofan engine, in particular to a method for predicting the remaining life of an aviation turbofan engine based on multi-source data fusion. Background technique [0002] Aeroengine prediction and health management technology (Prognostics and Health Management, PHM) provides technical guidance and prior knowledge for engine Condition-Based Maintenance (CM), and vigorously promotes the aero-engine maintenance technology gradually from scheduled maintenance to condition-based maintenance. The transformation of maintenance technology, ensuring flight safety and reducing maintenance costs has aroused extensive attention and research. With the advancement of sensor technology and electronic equipment, the original engine monitoring data can be accurately monitored and recorded, providing sufficient data support for PHM technology. How to accurately predict the remaining l...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the shortcomings of traditional aero-engine remaining life prediction methods, the present invention provides a multi-source statistical data-driven aero turbofan engine remaining life prediction method to solve the problem of ignoring engine multi-dimensional monitoring data and equipment operation in traditional statistical data-driven prediction methods environmental and individual differences

[0007] The basic concept of the present invention is to make full use of the effective monitoring data collected by the engine sensor, and perform information fusion on the basis of redundant features that are not conducive to life prediction to extract health indicators and failure thresholds that characterize the operating state of the engine, so as to solve the problem of traditional prediction methods. The problem of insufficient utilization of monitoring data information

Images