Mixed failure detection diagnosis method based on logical deduction and failure identification

Abstract

The invention provides a mixed failure detection diagnosis method based on logical deduction and failure identification. The method comprises the following steps of: acquiring a failure causal relationship and historical failure data of a system to be detected, establishing a failure tree of the system to be detected, and determining the reliable distribution of each event node of the failure tree; acquiring a plurality of latent failure modes of the system to be detected, respectively establishing failure models which correspond to a plurality of latent failure modes, and thus obtaining a plurality of latent failure models; establishing association relationships between the failure tree and a plurality of latent failure models of the system to be detected; and when the system to be detected fails, detecting the current failure mode, and thus obtaining a diagnosis result of the current failure mode according to the association relationship between the current failure model which corresponds to the current failure mode and the failure tree. By adoption of the method, failures of an aircraft can be accurately detected and diagnosed, specific parts which fail are found, the accuracy and identification capacity of failure detection diagnosis are improved, and foundation is provided for maintenance of the aircraft.

Description

technical field [0001] The invention relates to the field of aircraft fault detection and diagnosis, in particular to a hybrid fault detection and diagnosis method based on logical reasoning and fault identification. Background technique [0002] With the increasing integration, complexity, comprehensiveness and intelligence of modern aircraft, people pay more and more attention to the fault diagnosis, maintenance support and reliability of the system. Among them, the comprehensive fault detection and diagnosis technology is very important Improving the safety performance of aircraft and improving the efficiency of maintenance support play an important role. [0003] Now, fault detection and diagnosis technology is widely used in aerospace, automobile, metallurgy and other fields. Commonly used fault diagnosis methods include fault tree analysis, filter method, multi-model method, expert system, logical reasoning, neural network and other methods. Some of these methods are...

AI Technical Summary

Problems solved by technology

This method also has its disadvantages. It can determine the failure mode of the system, but it cannot go deeper to determine the faulty components.

For example, the airspeed information of the aircraft can be detected by using the multi-model method, but it is impossible to determine which device has failed, whether it is a pitot tube or an AC motor.

[0006] Therefore, different fault detection and diagnosis methods have their own deficiencies, and cannot perfectly solve the problem of fault detection and diagnosis.

Images