Radial Basis Function (RBF) Network Based Adaptive Tcps Controller for Power System

In this paper, the application of thyristor controlled phase shifter (TCPS) in damping power system oscillation is investigated. Analysis is carried out considering TCPS equipped with proportional-integral-derivative (P-I-D) controller. Gain settings of the P-I-D Controller for TCPS are tuned in real time using a radial basis function (RBF) neural network. The RBF based neural network is trained using an orthogonal least squares (OLS) learning algorithm. Dynamic performances considering gains tuned offline using GA and with the gains tuned online with RBF based neural network for TCPS equipped P-I-D controller are compared, and simulation results show that dynamic performance of the system with an RBF network based P-I-D controller for TCPS is virtually identical to that with the P-I-D controller gains tuned off-line using genetic algorithm (GA) for TCPS. It is also found that RBF based adaptive P-I-D controller for TCPS does not adversely affect the transient stability and damps out the oscillation following fault clearing.