A New Fault Diagnosis Model of Electric Power Grid Based on Rough Set and Neural Network
Liying Zhang, Wang Dazhi, Cuiling Zhang, Liu Xiao-qin · 2012
Fault diagnosis for system quick return to normal after the accident has important significance. On the basis of giving a new type of attribute reduction method, a coupling recognition model is established which combines rough set and neural network closely in this paper. It used rough set theory to get the most simple decision rules from the data samples, to guide to establish neural network structure. Using rough membership function initializes the network parameters, in order to reduce the network training iterative times and improve the network convergence speed. The simulation results illustrate that the model improves network's structure, and its recognizing effects are obvious and its classifying ability is strong, as well as the model is very error permissible and explicable. It has very wide foreground.