Artificial circuit fault diagnosis pattern sorting algorithm

Abstract

The invention discloses an artificial circuit fault diagnosis pattern sorting algorithm based on signal characteristic space modeling. According to the method, signals collected by test nodes are utilized, optimal fractional Fourier transform (FrFT) and R type cluster analysis are performed on the signals on the basis of the maximum entropy principle (MEP) to describe the characteristics of fault samples, and different spatial distribution modeling of faults are conducted; according to the sort separability criterion of the characteristic evaluation of minimum-in-cluster-distance and maximum-between-cluster-distance, objective optimization functions of nuclear parameters are constructed, and on the basis of a self-adaption genetic algorithm, the objective functions are optimally solved, and the nuclear parameters are adjusted; in combination with Q type cluster analysis, a hierarchical support vector machine classifier (SVC) is constructed to find and separate the faults; and through the algorithm, sensitivity reflecting fault characteristics can be extracted from measurement signals, and higher fault diagnosis speed and higher fault diagnosis accuracy are achieved. The fault diagnosis examples of a Continuous-Time State-Variable Filter circuit and an ML-8 radar prove the speediness and effectiveness of the algorithm.

Description

technical field [0001] The invention relates to a fault diagnosis method of an analog circuit, which belongs to the technical field of test diagnosis, realizes a pattern classification method through signal feature space modeling, can be used to judge the working state of the analog circuit part in a missile, and can diagnose faults state for discovery and separation. Background technique [0002] Fault diagnosis of analog circuits began in the 1960s, and its theoretical research started from the solvability of network element parameters. However, due to its unique difficulties such as diversity of fault states, tolerance of element parameters, insufficient information and Due to the complexity of the structural model, etc., the development of research on fault diagnosis of analog circuits is relatively slow, and its testing and fault diagnosis has always been a problem that plagues the circuit testing industry. After the 1990s, with the development of artificial intelligen...

AI Technical Summary

Problems solved by technology

[0002] Fault diagnosis of analog circuits began in the 1960s, and its theoretical research started from the solvability of network element parameters. However, due to its unique difficulties such as diversity of fault states, tolerance of element parameters, insufficient information and Due to the complexity of the structural model, the development of research on fault diagnosis of analog circuits is relatively slow, and its testing and fault diagnosis have always been a problem that plagues the circuit testing industry.

After the 1990s, with the development of artificial intelligence technology, fuzzy theory, wavelet technology and some machine learning methods have been applied in this field and achieved good results, but they all have one-sidedness and have no effect on solving actual analog circuit faults. There is still a certain gap between diagnosis and analysis of problems

Images