Fault detection method, system and device for liquid rocket engine and medium

Abstract

The invention discloses a fault detection method, system and device for a liquid rocket engine and a medium, and the method comprises the steps of constructing a first BP neural network model according to the structural features of the liquid rocket engine; optimizing the first BP neural network model by adopting an adaptive genetic algorithm to obtain an optimal weight and an optimal threshold; according to the optimal weight and the threshold value, building an AGABP network model; inputting historical data of the liquid rocket engine during normal work into the AGABP network model for training, and obtaining a second AGABP network model; inputting real-time data of the liquid rocket engine into the second AGABP network model, comparing an output result of the second AGABP network model with real-time detection data of a sensor in the liquid rocket engine, and judging whether the liquid rocket engine breaks down or not. According to the invention, real-time fault detection can be carried out on working of the liquid rocket engine, and compared with a BP neural network model prediction result, the method has higher system sensitivity and robustness.

Description

technical field [0001] The invention relates to the technical field of engine fault detection, in particular to a liquid rocket motor fault detection method, system, equipment and medium. Background technique [0002] Liquid rocket engine (Liquid Rocket Engine, LRE) is an important power device and component of the launch vehicle system. Its reliability and safety are necessary guarantees for the successful completion of aerospace tasks. Through effective fault diagnosis and health management of liquid rocket engines, It can reduce the probability of accidents and reduce unnecessary property losses. Currently, fault detection methods for liquid rocket engines are mainly divided into three categories: model-driven methods, data-driven methods, and artificial intelligence-based methods. [0003] At present, in the method of fault detection for liquid rocket engines, the method based on model needs to establish a system model according to the law of system operation, and the a...

AI Technical Summary

Problems solved by technology

However, artificial intelligence-based methods are currently mostly used in fault classification, and most of the artificial intelligence-based detection methods used in fault detection cannot meet the needs of real-time detection, the detection speed is slow, and the detection accuracy is low.

Images