Performance parameter based civil aircraft engine starting system health monitoring method

Abstract

The invention discloses a performance parameter based civil aircraft engine starting system health monitoring method and relates to the technical field of equipment state monitoring. On the basis of aircraft QAR data, a performance parameter baseline model of important parameters in a civil aircraft engine starting system is established with a data mining technology; then, a covariance matrix of test samples is calculated; finally, the similarity and the mahalanobis distance between the test samples and the baseline model are obtained with a mahalanobis distance technology, the health state ofthe system is represented with the distance, and the purpose of health monitoring is achieved. The method is applied to the civil aircraft engine starting system, the problem of insufficiency of performance evaluation means of the aircraft engine starting system is solved, the performance state of the engine starting system can be accurately evaluated, early warning is timely given at the early stage of failure of the engine starting system, and unplanned accidents are reduced; and besides, performance evaluation data come from existing civil aircraft QAR data and have higher practicability and operability.

Description

technical field [0001] The invention relates to the technical field of equipment state monitoring, in particular to a method for monitoring the health of a civil aircraft engine starting system based on performance parameters. Background technique [0002] Health monitoring is the basis for automatic fault diagnosis and condition-based maintenance of the aircraft engine starting system. It can not only greatly reduce unplanned maintenance, improve the economy and timeliness of aircraft maintenance; on-wing time. In view of the relevance and coupling of the accessory systems of the engine, the health monitoring of the starting system can also ensure the safe operation of the engine body and avoid the phenomenon of engine over-rotation and engine starting suspension due to the failure of the starting system. Systematic implementation of health monitoring has great practical significance. [0003] At present, on the route, due to the complexity and marginality of the starting...

AI Technical Summary

Problems solved by technology

However, airlines currently adopt a single visualization method without using relevant technologies of data mining, and cannot extract valuable information from massive operational data.

[0005] In summary, there is a lack of a non-parametric modeling method in the prior art to evaluate the performance of the engine starting syste

Images