System fault inference method

Abstract

The invention relates to a system fault inference method, and relates to the technical field of aviation fault testing. The method is based on a fault diagnosis expert system and a multi-signal flow graph testability module. The method specifically comprises the steps: carrying out the fault judgment of a diagnosed system under a single state parameter, identifying a judgment result through a double-attribute value, and finally forming a state matrix of the diagnosed system; carrying out the testability modeling of the diagnosed system, and obtaining a correlation matrix of the fault mode of the diagnosed system and the state of the diagnosed system; carrying out the calculation of the state matrix of the diagnosed system and the correlation matrix, and obtaining a diagnosis result of thediagnosed system. According to the invention, the method is high in speed and precision when the method is used in a fault inference process.

Description

technical field [0001] The invention belongs to the technical field of aviation fault testing, in particular to a system fault reasoning method. Background technique [0002] A large complex equipment system is a complex cross-linked system of mechanical, electrical, hydraulic, gas and control systems. There are many components and various types. The failure of a single component will cause multiple abnormalities or functional degradation of the system. The failure phenomenon is very confusing. It is not easy to judge the fault source component. When two or more components fail, the fault phenomenon is more complicated, and fault detection and isolation become more difficult; especially when the system has redundant configuration, function backup and reconstruction capabilities, or the complex system of multiple subsystems When the cross-links work together to complete the function, the fault may only trigger the local functional abnormality of the system, or may cause the ...

AI Technical Summary

Problems solved by technology

When initially creating and maintaining the knowledge base, domain experts need to manage the knowledge base through the man-machine interface to provide sufficient "knowledge" of system faults for the reasoning machine to diagnose faults. and technical level limitations, it is difficult to establish sound diagnostic rules in a short time

Once the fault diagnosis expert system scheme is adjusted, the diagnosis rules need to be re-analyzed and organized, and the technology development process is unstable and poorly adaptable

Moreover, in the fault diagnosis process, the reasoning machine also needs to decide whether to ask the user for new diagnostic symptom information according to the degree of diagnosis. The diagnosis process requires human-computer interaction, and it is difficult to realize online fault detection and diagnosis.

Images