{"id":849,"date":"2024-12-31T16:48:10","date_gmt":"2024-12-31T16:48:10","guid":{"rendered":"https:\/\/els-journal.net\/wp\/about\/"},"modified":"2025-04-27T08:29:47","modified_gmt":"2025-04-27T08:29:47","slug":"28203-analog-circuits-fault-diagnosis-using-ism-technique-and-a-ga-svm-classifier-approach","status":"publish","type":"page","link":"https:\/\/els-journal.net\/wp\/?page_id=849","title":{"rendered":"28203 Analog Circuits Fault Diagnosis Using ISM Technique and a GA-SVM Classifier Approach"},"content":{"rendered":"\n\n\n

Vol. 28, No. 2 – December 2024<\/h3>\n

Analog Circuits Fault Diagnosis Using\nISM Technique and a GA-SVM Classifier Approach<\/h3>\n

https:\/\/doi.org\/10.53314\/ELS2428054K<\/h5>\n
Sabah Kouachi, Nacerdine Bourouba, Kamel Mebarkia, and Imad Laidani<\/h5>\n
Abstract<\/b><\/h5>\n
This present work aims to contribute to the solution\nof the problems encountered in electronic circuits fault diagnosis.\nOne of these troubleshoots faced is the lack of effective features\nthat help to optimize fault classifier and hence improve circuit\nfault detection and identification. Thus, our feature extraction\napproach is based on the CUT\u2019s transfer function. This is deduced\nfrom the Matlab identification system IS model (ISM), namely the\nOE model belonging to the ARMA model\u2019s family. These features\nare the transfer function polynomial coefficients playing a crucial\nrole in the fault free and faulty circuits construction models and\nfeeding the classifier for the fault diagnosis purpose. The faults\nwe are dealing with are of single parametric type. This is done\nfrom PSPICE time domain analysis on the CUT output response\nunder theses circuit conditions and followed by extracting the IS\nmodel (ISM) orders (p,q) polynomials. The coefficient values of the\nlatter were considered as efficient comparison elements between\nfaulty and healthy circuit responses. As a result, the OE model has\nachieved 100% fault coverage and its construction reached high\naccuracy level exceeding 98% for faulty circuits. This accuracy\nlevel ambition us to use its coefficients as input features for our\nHybrid proposal fault classifier. This is built with GA and SVM\nalgorithms combination targeting both data reduction and fault\nclassification accuracy respectively. The results achieved are conclusive\nsince the classifier accuracy level reached 100% and a 70%\nof feature data volume reduction was scored.<\/h5>\n
Full text: <\/a><\/h5>\n\n\n\n\n\"cited<\/a>\n\n\n\n\nGoogle Scholar Citations <\/a>\n\n\n\n\n
<\/span> <\/center>