Dipayan Guha Ph.D Scholar, EE Department NIT-Durgapur, Durgapur West Bengal, India guha.dipayan@yahoo.com Provas Kumar Roy H.O.D, EE Department Dr. B.C.Roy Engineering College Durgapur, West Bengal, India roy_provas@yahoo.com Subrata Banerjee Professor, EE Department NIT-Durgapur, Durgapur West Bengal, India bansub2004@yahoo.com Nomenclature: ACE = Area Control Error B i = Frequency bias constant DB = Dead Band D i = P Di / f i = Load frequency constant f = Nominal System frequency H i = Inertia constant i = Subscript referring to area (i = 1, 2) J = Objective function or cost index K P , K d , K i = Electric Governor’s proportional, derivative and integral gains, respectively K pi = 1/D i = Gain of power system P ri = Rated Power (a 12 = – P r1 / P r2 ) R i = Speed regulation parameter T 12 = Synchronizing Coefficient T g = Steam governor time constant K r = Steam turbine reheat constant T r = Steam turbine reheat time constant T t = Steam turbine time constant T pi = 2H i /f*D i =Time constant of power system T w = Water starting time P tie = Incremental change in tie-line power f i = Incremental frequency deviation P Gi = Incremental generation change X Ei = Incremental change in governor valve position P Di = Incremental load change Abstract — This article proposes automatic generation control (AGC) of an interconnected three equal and unequal hydro- thermal system with DB non-linearity. Moreover, the self tuning control scheme of superconducting magnetic energy storage unit (SMES) is performed to investigate the performances of AGC problem. Dynamic responses of SMES connected AGC are compared with that of integral (I) and proportional-integral- derivative (PID) controlled AGC. Frequency deviation signal is used as an input to SMES. Integral square error approach with Biogeography based optimization algorithm is used to find optimum values of controller parameters. 1% step load perturbation in either area is considered for simulation study. Simulation study exhibits significant effect of designed SMES based controller on the dynamic performances of an interconnected power system with sudden load perturbation. Index Terms — Area control error, automatic generation control, biogeography based optimization, dead band, integral square error, superconducting magnetic energy storage I. INTRODUCTION odern power system networks comprising number of power system utilities connected together through transmission line called as Tie-line. The successful operation of power system network is to match total generation with total load demands plus losses associated with the system. Any sudden change in load causes deviation of system frequency and tie-line power from their nominal values. Especially, if frequency of changing loads is in the vicinity of inter-area oscillation modes (0.2 to 0.8 Hz), system frequency may be heavily disturbed and oscillate. Large oscillation causes serious stability problem of an interconnected power system [1]. Under this situation, conventional governor control is no longer able to compensate such oscillations hence secondary or supplementary control scheme is added with governor control for improving dynamic responses of an interconnected power system. The function of AGC can be viewed as supplementary control function which attempts to matches generation with randomly changing loads [2]. As per the IEEE standards, AGC is the any supplementary control that automatically adjusts output power levels with generation within a control area [3]. The main objective of AGC is to restore system frequency and tie-line power to their scheduled values. Many investigations in the field of AGC of an isolated or interconnected power system have been reported in past. Literature survey shows that little works have been reported on multi-area hydrothermal power system [2, 4, 5, 6]. A number of control strategies such as classical, genetic algorithm (GA), fuzzy logic, particle swarm optimization (PSO), artificial neural network (ANN) etc. are proposed to improve dynamic performances of AGC based power system, have been reported in [4]-[14]. Training of these controllers is not an easy task. It requires appropriate learning algorithm, availability sufficient and adequate learning data, suitable number of neurons etc., which increases with complexity and size of power system unit. Hence, it is quite difficult to implement in practice. The internal characteristic of power system network is highly non-linear; hence optimal values of controller parameters are not sufficient to give better dynamic performances of AGC system. Therefore, to add more damping on the system oscillation, small size power system stabilizer (PSS) is connected to AGC. Tuning of stabilizer parameters is very important for AGC study; otherwise it produces destabilizing effect to the power system network. Superconducting magnetic energy storage unit (SMES) is more capable of Optimal Design of Superconducting Magnetic Energy Storage Based Multi-Area Hydro-Thermal System Using Biogeography Based Optimization M 2014 Fourth International Conference of Emerging Applications of Information Technology 978-1-4799-4272-5/14 $31.00 © 2014 IEEE DOI 10.1109/EAIT.2014.27 52