Fake Measurement Detection in Digitalized Substation Automation System using EA Based RBFNN

With the development of communication and information technology, electronic instrumental transformer (EIT) and Ethernet based communication network has got prevalent in digitalized substation automation system. Since malfunction of EIT can produce dangerous fake measurements, the risk for the protection system to receive these fake measurements and mis-operate increase. Pattern identification can be utilized to detect fake measurements. In order to achieve satisfying pattern identification precision within time limit imposed by protection systems, evolution algorithm (EA) has been utilized to optimize construction of radial basis function neural network (RBFNN) in the paper. EA algorithm is used to prune network scale in order to mitigate contradictory requirements of high precision and low time delay. Simulation results show EA based RBFNN can achieve satisfying performance within limited time.