Article Preview
TopIntroduction
The major objectives of the interconnected massive power system is to generate, distribute and regulate the electric energy for providing reliable and quality power to the consumers. Due to rapid growth in demand of electricity, structure of power system is getting more complex, which needs proper plan and design for secure and reliable operation in stringent loading conditions. Power systems are usually exposed to large number of disturbances during their normal operation. As an immediate consequence, the real and reactive power balance between generation and demand gets affected. Voltage profile, frequency, and tie-line power flow which are some of the important measures of power quality, get deviated from their usual steady state values. The real power mismatch mainly causes the deviation in frequency and tie line power exchange between different areas of the system whereas, fluctuation in voltage profile is due to the reactive power mismatch (Kundur 2009). Load frequency control (LFC) plays a vibrant role in maintaining the frequency and tie line power flow within a predefined value at steady state. It continuously monitors the deviation in frequency and tie-line power exchange to calculate the change in generation required, usually referred as area control error (ACE) (Kundur 2009; Chon 1956).
Many researchers have been suggested several control strategies to maintain the frequencies of different areas and tie-line power flow between them within a predefined value at steady state during normal/disturbed operating condition, since last few decades. For the application of power industry, PI and PID controllers are more popular because of their advantages like ease to implement and low cost. Golpira and Bevrani (2011)was employed genetic algorithm (GA) for LFC considering three area model of power system (each area having a thermal unit) incorporating different types of physical constraints. Ghosal (2004)concluded that particle swarm optimization (PSO) is faster as compared to GA and GA-SA (simulated annealing), although their transient performances are more or less equal for a three area thermal power system. The performance of single reheat against double reheat turbine was compared by (Nanda, Mangala and Suri 2006) for a two-area thermal-hydro (with mechanical & electric governor) power system including generation rate constraint (GRC) to make the study more realistic. Gozde, Taplamacioglu and Kocaarslan (2012) implemented artificial bee colony (ABC) to tune the parameters of PI/PID controller and compared with PSO technique for LFC with different objective functions. The bacterial foraging optimization (BFO) algorithm based optimized PID controller was introduced by (Ali and Abd-Elazim 2013) and compared with GA &Ziegler Nichols (ZN) for two-area thermal system (without reheat turbine) incorporating different values of GRC. According to Sahu, Panda and Pradhan (2015) the hybrid firefly algorithm-Pattern Search (hFA-PS) based PID controller performs better over BFO, GA, & ZN controller for LFC. Mohanty, Panda and Hota (2014) used differential evolution (DE) optimized PI/PID controller in two area multi-source (thermal, hydro, wind & diesel) system for AGC to make it more realistic. Kouba,Menaa, Hasni and Boudour (2017] proposed a PID controller designed with hybrid GA-PSO algorithm for optimal LFC study. Teaching learning based optimization (TLBO) algorithm tuned PIDD controller was proposed by (Sahu, Sekhar and Panda 2016) initially for two area thermal power system model (PSM) and further extended for multi-source system considering suitable nonlinearities. Mohanty and Hota (2015) implemented fruit-fly algorithm based several classical controller for two-area thermal-hydro-nuclear power system in a deregulated environment. The two degree of freedom-PID (2DOF-PID) controller is having more flexibility to enhance the performance due to the presence of two more extra controlling parameters which are absent in conventional PID controller. Sahu, Panda and Rout (2013)used 2DOF-PID controller designed using DE technique for LFC in a thermal system considering modified objective function and different nonlinearities.Several 2DOF controllers, optimized using cuckoo search algorithm was considered by (Dash, Saikia and Sinha 2014) to verify their performance with different conditions for AGC.