TY - GEN
T1 - Neural networks on transformer fault detection
T2 - 2011 IEEE/PES Power Systems Conference and Exposition, PSCE 2011
AU - Msiza, I. S.
AU - Szewczyk, M.
AU - Halinka, A.
AU - Pretorius, J. H.C.
AU - Sowa, P.
AU - Marwala, T.
PY - 2011
Y1 - 2011
N2 - Following a number of studies that have employed different forms of neural network models to perform dissolved gas-in-oil analysis (DGA) of transformer bushings, this manuscript focuses on evaluating the relevance of the parameters that form part of the model input space. Using a multilayer neural network initially populated with all the 10 input parameters (10V-Model), a matrix containing causal information about the possible relevance of each input parameter is obtained. The information from this matrix is proven to be valid through the construction and testing of another two, separate, multilayer networks. One network's input space is populated with the 5 most relevant parameters (MRV-Model), while the other is populated with the 5 least relevant parameters (LRV-Model). The obtained classification accuracy values are as follows: 100% for the 10V-Model, 98.5% for the MRV-Model, and 53.0% for the LRV-Model.
AB - Following a number of studies that have employed different forms of neural network models to perform dissolved gas-in-oil analysis (DGA) of transformer bushings, this manuscript focuses on evaluating the relevance of the parameters that form part of the model input space. Using a multilayer neural network initially populated with all the 10 input parameters (10V-Model), a matrix containing causal information about the possible relevance of each input parameter is obtained. The information from this matrix is proven to be valid through the construction and testing of another two, separate, multilayer networks. One network's input space is populated with the 5 most relevant parameters (MRV-Model), while the other is populated with the 5 least relevant parameters (LRV-Model). The obtained classification accuracy values are as follows: 100% for the 10V-Model, 98.5% for the MRV-Model, and 53.0% for the LRV-Model.
KW - fault detection
KW - neural networks
KW - parameter relevance
KW - transformer bushings
UR - http://www.scopus.com/inward/record.url?scp=79958775449&partnerID=8YFLogxK
U2 - 10.1109/PSCE.2011.5772567
DO - 10.1109/PSCE.2011.5772567
M3 - Conference contribution
AN - SCOPUS:79958775449
SN - 9781612847870
T3 - 2011 IEEE/PES Power Systems Conference and Exposition, PSCE 2011
BT - 2011 IEEE/PES Power Systems Conference and Exposition, PSCE 2011
Y2 - 20 March 2011 through 23 March 2011
ER -