Detection and Classification of Transmission Line Congestion by Feed Forward and Radial Basis Function Neural Networks

AbstractA novel scheme for detection and classification of transmission line congestion in power systems using feed forward and radial basis function neural networks is presented. In the detection stage, the neural network is trained to learn the complex mapping between various nonlinear multi-objective input functions and the specified target in order to arrive at the desired goal of identifying the lines being overloaded or congested. The cases indicating line congestion during the detection stage are presented to the radial basis function neural network for classification of the congested lines as per their degree of severity. The results obtained from MATLAB-based case study conducted on the IEEE 30-bus test system infer that the proposed feed forward neural network with Levenberg–Marquardt backpropagation scheme offers the fastest convergence as compared to other types of backpropagation schemes and the conventional approach as well. Further, the radial basis function neural network shows convincingly faster convergence during the severity classification stage.