Title :
A two stage neural network DC fault dictionary
Author :
Rutkowski, Jerzy
Author_Institution :
Inst. of Electron., Silesian Tech. Univ., Gliwice, Poland
fDate :
30 May-2 Jun 1994
Abstract :
A new concept of neural network DC fault dictionary is presented. The neural network is combined with two networks: Learning Vector Quantization network and Multilayer Perceptron. Simulation results are presented and they show that the proposed approach is an improvement as compared with classical nearest neighbour rule approach
Keywords :
analogue circuits; fault diagnosis; learning (artificial intelligence); multilayer perceptrons; neural nets; pattern classification; vector quantisation; DC fault dictionary; analogue circuits; fault isolation; fault location; learning vector quantization network; multilayer perceptron; neural network; simulation before test; Circuit faults; Circuit simulation; Circuit testing; Dictionaries; Euclidean distance; Fault detection; Multilayer perceptrons; Neural networks; Vector quantization; Voltage;
Conference_Titel :
Circuits and Systems, 1994. ISCAS '94., 1994 IEEE International Symposium on
Conference_Location :
London
Print_ISBN :
0-7803-1915-X
DOI :
10.1109/ISCAS.1994.409585
