Combination Neural Network Circuit Fault Diagnosis Method Considering Fuzzy Group Pre-discrimination

Abstract

The invention discloses a combined neural network circuit fault diagnosis method considering fuzzy group pre-discrimination, and belongs to the technical field of fault diagnosis. The method first determines the fault isolation fuzzy group of a test point to perform separability determination of the fault fuzzy group, establishes a combined neural network diagnosis sub-model for the separable fuzzy group state-test matrix, and uses the combined neural network to complete the fault diagnosis. The method provided by the invention makes the setting of classification labels in the neural network model more reasonable, improves the accuracy of the fault diagnosis based on the neural network, and has high operation efficiency by running at most one neural network model for each diagnosis, whichexpands the application range of the neural network model, and improves the diagnostic effect.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis and relates to a method for fault diagnosis of a combined neural network circuit considering fuzzy group pre-discrimination. Background technique [0002] The failure of the product will affect the system function, cause the system to fail and even cause a major accident. The study of fault diagnosis technology is of great significance to improve system efficiency, reduce downtime, and reduce safety hazards. The fault diagnosis method based on neural network has the ability of self-learning and self-adaptation, and has been widely used in aviation, aerospace, shipbuilding, automobile and other fields. [0003] However, in engineering practice, for products with a large number of failure modes and test parameters, it is very difficult to use only one neural network for fault diagnosis, and the accuracy of the model will decrease with the increase of the number of faults. Establishing a di...

AI Technical Summary

Problems solved by technology

[0003] However, in engineering practice, for products with a large number of failure modes and test parameters, it is very difficult to use only one neural network for fault diagnosis. The accuracy of the model will decrease with the increase of the number of faults, and for each fault Establishing a diagnostic model will lead to waste of resources and affect diagnostic efficiency

Images