Adaptive fuzzy logic load frequency control of multi-area power system Hassan Yousef Department of Electrical and Computer Engineering, Sultan Qaboos University, Muscat, Oman article info Article history: Received 20 December 2013 Received in revised form 22 December 2014 Accepted 24 December 2014 Available online 17 January 2015 Keywords: Adaptive Fuzzy control Load frequency Multi-area abstract Based on indirect adaptive fuzzy control technique, a new load frequency control (LFC) scheme for multi- area power system is proposed. The power systems under study have the characterization of unknown parameters. Local load frequency controller is designed using the frequency and tie-line power deviations of each area. In the controller design, the approximation capabilities of fuzzy systems are employed to identify the unknown functions, formulate suitable adaptive control law and updating algorithms for the controller parameters. It is proved that the proposed controller ensures the boundedness of all vari- ables of the closed-loop system and the tracking error. Moreover, in the proposed controller an auxiliary control signal is introduced to attenuate the effect of fuzzy approximation error and to mitigate the effect of external disturbance on the tracking performance. Simulation results of a three-area power system are presented to validate the effectiveness of the proposed LFC and show its superiority over a classical PID controller. Ó 2014 Elsevier Ltd. All rights reserved. Introduction Large scale power system consists of number of control areas and each area represents coherent group of generators. The control areas are interconnected through tie-lines. Each control area is equipped with primary and supplementary control actions. Depending on the governor droop characteristic and frequency sen- sitivity of the load, the primary control action provides steady-state frequency deviation for a change in system load. Restoration of sys- tem frequency to nominal value requires supplementary control action that adjusts the load reference set point through the speed-changer motor. Therefore, the basic means of controlling prime-mover power to match variations in system load is through control of the load reference set points of selected generating units [1]. The main objectives of the LFC are to regulate the system fre- quency to the specified nominal value and to maintain the inter- change power between control areas at the scheduled levels. A survey of different control schemes of LFC can be found in [2,3]. Many controllers have been proposed for power system LFC problems in order to achieve a satisfactory dynamic performance. The most widely employed controller is the fixed gain type, like a PI or a PID controller. A new control structure with a tuning method to design a PID load frequency controller for power sys- tems is presented in [4]. Implementation of the LFC requires accu- rate information about the control area parameters, which are usually imprecisely modeled or varying due to wearing out of the components. To overcome this difficulty, intelligent control techniques have been used. The last two decades have witnessed increasing attention for applications of intelligent techniques such as fuzzy systems, artificial neural networks, genetic algorithms, etc. to deal with several aspects of power systems [5]. Adaptive fuzzy output tracking excitation control of power system generator is presented in [6]. In [7], application of a fuzzy gain scheduled proportional and integral controller for load-frequency control of two-area power system is presented. A control scheme based on artificial neuro-fuzzy inference system (ANFIS) is proposed in [8] to optimize and update control gains for automatic generation con- trol (AGC) according to load variations. A fuzzy system is used in [9] to determine adaptively the proper proportional and integral gains of a PI controller according to the area-control error and its change for LFC. The LFC for power system subject to nonlinearities in valve position limits and parametric uncertainties is developed using T-S fuzzy system [10]. A method based on type-2 fuzzy sys- tem for load frequency control of a two-area interconnected reheat thermal system including superconducting magnetic energy stor- age (SMES) units is proposed in [11]. A Genetic Algorithm (GA) based fuzzy gain scheduling approach for load frequency control is presented in [12]. Two robust decentralized control design methodologies for LFC using linear matrix inequalities and genetic algorithms optimization are proposed [13]. Bacterial Foraging Optimization Algorithm (BFOA) is employed [14] to search for opti- mal LFC PID controller parameters to minimize the time domain objective function. A decentralized model predictive load fre- quency control was developed for multi-area power system http://dx.doi.org/10.1016/j.ijepes.2014.12.074 0142-0615/Ó 2014 Elsevier Ltd. All rights reserved. E-mail addresses: hyousef@squ.edu.om, hyousef456@yahoo.com Electrical Power and Energy Systems 68 (2015) 384–395 Contents lists available at ScienceDirect Electrical Power and Energy Systems journal homepage: www.elsevier.com/locate/ijepes