A Neural Network Approach to Design Power System Stabilizer for Damping Power Oscillations

Modern power system networks are structurally complex and are prone to several undesired phenomena like outage of lines and generators, transmission line faults, power oscillations etc. The power oscillations exist in the system after disturbances when generators swing with respect to each other. However, these oscillations are required to be damped sufficiently to ensure reliable operation of system. Consequently, in this work, a neural network approach is adopted to design power system stabilizer (PSS) for damping power oscillations in a single-machine infinite bus test system. For neural network, a radial basis function neural network (RBFNN) is chosen to train the parameters of PSSs. Diverse test cases are considered to test the performance of proposed PSS and the results are compared with results obtained from conventional PSSs. The performance of proposed PSS is found to be efficient.