Intelligent fault diagnosis method of aero-engine control system sensor based on mode gradient spectral entropy

Abstract

The invention provides an intelligent fault diagnosis method of an aero-engine control system sensor based on a mode gradient spectral entropy, and the method comprises the steps: firstly collecting data of a plurality of measurable sensors in an aero-engine control system in different working states, a normal state and different fault states of the sensor of an aero-engine, and forming a sample data set; preprocessing collected sensor data, and processing preprocessed sample data of each health state through a mode gradient spectral entropy method to obtain a corresponding spectral entropy graph; taking the spectral entropy graph as input to train a CNN network; and after the trained CNN model is obtained, carrying out preprocessing and PGSE analysis on real-time measurement data of the aero-engine control system sensor to obtain a spectral entropy graph, and obtaining a sensor fault diagnosis result by using the CNN model. According to the method, the sensor fault diagnosis accuracy can be improved, the fault mode of a complex nonlinear aero-engine system can be efficiently and accurately recognized, the robustness requirement is met, and the method is not sensitive to noise interference.

Description

technical field [0001] The invention belongs to the field of fault diagnosis of an aero-engine control system, and in particular relates to an intelligent fault diagnosis method for a sensor of an aero-engine control system based on a mode gradient spectrum entropy and a convolutional neural network. Background technique [0002] Aero-engine control system is a complex thermomechanical machine working under high-speed, high-load, and high-temperature environmental conditions for a long time. As the heart of an aircraft, an aero engine plays a vital role and has high safety requirements. Since the aircraft engine control system relies on the measurement data of the sensor to work, if the sensor fails, the result will be disastrous. Therefore, detecting, isolating and regulating the sensor failure of the aero-engine is the key to improving its reliability. [0003] The complex structure of aero-engines leads to complex signal transmission paths and noise coupling, which make...

AI Technical Summary

Problems solved by technology

However, this method is unfounded and unsatisfactory, and the performance of the training model is poor, which cannot meet the requirements of rapidity and high precision for modern fault diagnosis. A faster and more reliable automatic diagnosis process is still needed.

Images