Artificial neural network PI controlled superconducting magnetic energy storage, SMES for augmentation of power systems stability

Abstract

This paper aimed to apply artificial neural network proportional, plus integral, PI controlled superconducting magnetic energy storage SMES to improve the transient stability of power systems. The PI controller parameters is firstly determined based on eigenvalue assignment approach. The artificial neural network, ANN is used to determine the optimum gains of the PI controller at different load values. The ANN is trained off line using Matlab software to obtain the optimum parameters of the PI controller. The speed deviation, Deltaomega and load angle deviation Deltadelta are used as input signal to the PI controller. The studied power system consists of single machine connected to an infinite bus via double transmission lines. The studied system is modeled by a set of nonlinear differential and algebraic equations and simulated by the Matlab software. The simulation results indicates the effect of the proposed ANN PI controlled SMES.