Research of RBF Neural Network Approach for Single Phase Grounding Fault of Distribution Network
Miao Qian · Electrical Measurement & Instrumentation · 2013
This paper introduces the main problem of single phase grounding fault location of distribution network,and proposes a new fault location method with RBF(radial basis function)Neural Network.The ability of feature extraction and pattern recognition is the characteristics of RBF Neural Network,and it is applied to analyze reflected wave signals in different branches.And with the C-type traveling wave location method,the purpose of precise location can be achieved.By setting up RBF network prediction model and comparing with forecasting method of BP,BP Neural Network could not judge the result correctly if the line model is complex and has more branches.The simulation results show that the RBF Neural Network is effective and superior to BP,and it lays the foundation for researching this method in the future.