Switching circuit fault classifying method based on wavelet transform and ICA feature extraction

Abstract

The invention discloses a switching circuit fault classifying method based on wavelet transform and ICA feature extraction. The method comprises the following steps: (1) generating a pseudo random signal as a test stimulation signal; (2) defining a fault mode; (3) acquiring the original response data of the circuit; (4) pre-treating the original response data by a Haar wavelet orthogonal filter; (5) extracting the fault feature parameters, and calculating the entropy and kurtosis as well as fuzzy sets thereof of low-frequency approximate information and high-frequency detail information for the pre-treated signal respectively; and (6) constructing a fault dictionary based on the extracted fault feature parameters so as to realize fault classification of the switching circuit. The method disclosed by the invention has the advantages of skillful concept, easiness in implementation and simulation proof and can distinguish the fault types more accurately than the existing method.

Description

technical field [0001] The invention relates to a switching circuit fault classification method based on wavelet transform and ICA feature extraction. Background technique [0002] Switched Current (SI) technology is an analog sampling data signal processing technology proposed in the late 1980s. As an alternative to switched capacitor technology, it processes continuous-time analog signals with discrete-time sampling data. It has the advantages of low voltage, high speed, low power consumption, small chip area and good high-frequency characteristics. It has gained rapid development in the past ten years. development of. SI technology does not contain linear capacitors and high-performance operational amplifiers, is fully compatible with digital CMOS process technology, and is easy to realize monolithic integration of large-scale digital-analog hybrid circuits. However, testing and diagnostics of the analog portion of mixed-signal circuits has been slow. Although many res...

AI Technical Summary

Problems solved by technology

Due to the measurement of current parameters, the amount of relevant fault information used for testing and diagnosis is incomplete, so that fault location cannot be accurately performed

[0004] In the literature [Guo, J., R., He, Y.G., Liu M.R..Wavelet neural network approach for testing of switched-current circuits.J Electron Test, 27: 611-625, 2011.] [recorded as literature [3 ]], the author uses the wavelet neural network to diagnose SI circuits and can correctly diagnose all hard faults, but the diagnosis rate for soft faults, especially for low-sensitivity transistors, is very low, only about 80%

In addition, in the literature [Long, Y., He, Y.G., & Yuan, L.F.Fault dictionary based switched current circuit fault diagnosis using entropy as a preprocessor.Analog Integrated Circuits and Signal Processing, 66(1), 2011:93-102.] [Recorded as document [1]], the author introduced the concept of fault feature preprocessing in SI circuit testing and diagnosis for the first time, through the information entropy preprocessing feature extraction of the collected fault response signals, and the calculation of information entropy fuzzy set construction Fault dictionary, for fault classification, reduces calculation and fault diagnosis time, and the diagnosis accuracy rate reaches about 95%, but this method is only suitable for fault diagnosis of small and medium-scale switching current circuits

In the literature [Zhang, Z., Duan, Z., Long, Y., & Yuan, L.F.A new swarm-SVM-based fault diagnosis approach for switched current circuit by using kurtosis and entropy as a preprocessor.Analog Integrated Circuits and Signal Processing, vol.81, no.1.2014.] [recorded as literature [2]], the author added a characteristic parameter - kurtosis, proposed the SI circuit test and Diagnosis method, the correct rate of soft fault diagnosis has been further improved, reaching about 99%, but it still cannot reach 100% fault diagnosis and identification

Images