Improved stochastic fractal search algorithm and modified cost function for automatic generation control of interconnected electric power systems

Abstract

An improved stochastic fractal search algorithm (ISFS) and a modified cost function are proposed in this paper to skillfully handle the issue of automatic generation control (AGC) of power systems. Most employed power system models namely two-area non-reheat thermal power system with and without governor dead band nonlinearity, and three-area hydro-thermal power plant with generation rate constraints are considered to be controlled by a PID controller. Then the gains of this controller are optimized with SFS and ISFS individually by minimizing the value of cost function proposed. This function consists in minimizing the integral time absolute error (ITAE) and also the time rates of frequency and tie-line power deviations. After recognizing the supremacy of SFS tuned PID controller over some existing methods in improving settling time and oscillations of frequency and tie-line power deviations, ISFS tuned PID controller is shown to promote the system performance further to compete with some other control schemes of higher degree and complexity available in the literature. This outcome has unveiled the superior tuning ability of ISFS over the original version of SFS. Also, convergence curves of the algorithms are analyzed from which it is observed that the speed of convergence for ISFS is remarkable.