Neural network-based aero-engine sensor fault self-diagnosis method

Abstract

The invention discloses an aero-engine sensor fault self-diagnosis method based on a neural network, and belongs to the field of aero-engine sensor fault diagnosis, and the method comprises the following steps: building an aero-engine component-level model, and collecting a sensor historical parameter database; designing a probability distribution function prediction model based on the neural network; designing a time sequence fault classification model based on the neural network; and in combination with the probability distribution mutation prediction model and the sensor fault classification model, constructing a sensor fault self-diagnosis framework, and realizing sensor fault self-diagnosis of the turbofan engine by adopting the constructed fault self-diagnosis framework. The method aims at solving the problem of fault diagnosis under the condition that a sensor can only access local historical data of the sensor, probability estimation is achieved through a neural network to solve the problems of uncertainty and randomness under the condition that information is incomplete, and feature extraction is conducted on time sequence data of the sensor to achieve fault classification.

Description

technical field [0001] The invention relates to a neural network-based fault self-diagnosis method for aero-engine sensors, which belongs to the field of aero-engine sensor fault diagnosis. Background technique [0002] With the rapid development of aviation propulsion technology, modern flight missions put forward higher requirements for the accuracy and stability of the engine control system. The full authority digital electronic control system (FADEC) Precision and relatively low maintenance costs have gradually replaced mechanical hydraulic control and become the mainstream trend of current aero-engine control systems. The digital electronic control system is composed of a large number of electronic components, sensors and actuators, and the aircraft engine is often in a high-temperature, high-pressure, complex and changeable working environment, which makes these components prone to failure during engine operation. Among them, the failure of the sensor It is the main r...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to solve the self-diagnosis problem of aero-engine sensors when they can only access local historical data. By considering the uncertainty caused by incomplete information, the probability of sensor parameters at the next moment is constructed according to historical data. Density distribution function, and realizes real-time judgment of mutation faults based on probability, and introduces time series fault classification model to realize judgment and classification of faults (hard failure, noise, drift) under long-term sequence, so as to ensure safe and reliable operation of the engine

Images