Solid-liquid rocket engine fault diagnosis method based on improved Bayesian algorithm

Abstract

The invention discloses a solid-liquid rocket engine fault diagnosis method based on an improved Bayesian algorithm, and belongs to the field of fault diagnosis, and the method specifically comprises the steps: firstly, for a solid-liquid rocket engine, adding fault factors to a part model, and obtaining various fault signals; fault signals are processed based on a stepping method, and observation information is improved from a one-dimensional space to a high-dimensional space; performing dimension reduction processing on the high-dimensional features by using a K-PCA method, and extracting principal component features of each fault signal; then, establishing an FCM-based fuzzy polymorphic Bayesian network, and training by using the principal component features to obtain the state number of each node of the Bayesian network; and finally, the state number of each node is utilized to correspond to fault probability distribution of each component model, the fault probability distribution is compared with respective set threshold values, and whether the corresponding component model has a fault or not is judged. The method achieves the fuzzy processing of the observation signal scale, improves the diagnosis capability of uncertain faults, and has the advantages of high accuracy and high adaptability.

Description

technical field [0001] The invention belongs to the field of fault diagnosis, and relates to a fault diagnosis method for a solid-liquid rocket motor, in particular to a solid-liquid rocket motor fault diagnosis method based on an improved Bayesian algorithm. Background technique [0002] The solid-liquid rocket engine has the incomparable advantages of liquid and solid engines in the comprehensive performance of structural complexity and thrust controllability, and has shown a wide range of application prospects in various aerospace applications. Judging from the current development of solid-liquid rocket engines at home and abroad, although their exploratory development has achieved certain success, reliability is still an important factor restricting their widespread use. Therefore, designing a highly reliable fault diagnosis system for solid-liquid rocket engines is of great engineering significance for its future practical application. [0003] The fault diagnosis of a...

AI Technical Summary

Problems solved by technology

[0004] Due to the complexity of the solid-liquid rocket engine, it is difficult to obtain an accurate mathematical model, so that the method based on the mechanism model often has low diagnostic accuracy in practical applications, or can only diagnose a few types of faults

The basis of learning and diagnosis based on neural network is experimental data. The solid-liquid rocket engine will lead to different experimental results due to different models and environments, and the cost of a single test is relatively high, and it is difficult to obtain a large amount of test data, so this method is not applicable. solid-liquid rocket motor

Images