Article Preview
TopIntroduction
The task of unit commitment (UC) involves scheduling of the generating units for minimizing the overall cost of the power generation over the scheduled time horizon while satisfying a set of system constraints. UC problem is a nonlinear, combinatorial optimization problem. The global optimal solution can be obtained by complete enumeration, which is not applicable to large power systems due to its excessive computational time requirements (Wood & Wollenberg, 1996). So far many methods have been developed for solving the UC problem such as priority list methods (Burns & Gibson, 1975; Sheble, 1990), integer programming (Dillon, Edwin, Kochs, & Taud, 1978; Garver, 1963), dynamic programming (DP) (Lowery, 1983; Ouyang & Shahidepour, 1991; Snyder, Powel, & Rayburn, 1987), branch-and bound methods (Cohen & Yoshimura, 1983), mixed-integer programming (Muckstadt & Wilson, 1986). These methods have only been applied to small UC problems and have required major assumptions which limit the solution space (Sen & Kothari, 1998; Sheble & Fahd, 1994). Lagrangian Relaxation (LR) (Fisher, 1981; Merlin & Sandrin, 1983; Zhuang & Galiana, 1988) can be applied to large UC problems due to its faster computational time. However, it suffers from numerical convergence and solution quality problems in the presence of identical units. Furthermore, solution quality of LR depends on the method to initialize and update Lagrange multipliers (Dekrajangpetch, Sheble, & Conejo, 1999).
In addition to the above methods, there is another class of numerical techniques applied to the UC problem. These are (GA) (Kazarlis, Bakirtzis, & Petridis, 1996), evolutionary programming (EP) (Juste, Kita, Tanaka, & Hasegawa, 1999) and simulated annealing (SA) (Zhuang & Galiana, 1990). These are general-purpose searching techniques based on principles inspired from natural systems. These methods have the advantage of accommodating more complicated constraints and are claimed to have better solution quality.
Ant colony optimization (ACO) was proposed by M. Dorigo et al to solve the difficult combinatorial optimization problems. ACO is a random stochastic population based algorithm that simulates the behavior of ants for cooperation and learning in finding shortest paths between food sources and their nest (Bonabeau, Dorigo, & Theraulaz, 1999; Dorigo, Mahiezzo, & Colorni, 1996; Dorigo & Gambardella, 1997; Dorigo, Di Caro, & Gambardella, 1999). In ACO, the ants’ behavior is simulated to solve the combinatorial problems such as traveling salesman problem and quadratic assignment problem (Dorigo, Mahiezzo, & Colorni, 1996; Dorigo & Gambardella, 1997). The ACO is applied to solve the UC problem by Simon, Padhy, and Anand (2006) and Yu, Chou, and Song, 1998).
This paper proposes a new method, Evolving Ant Colony optimization (EACO) for solving UC problem for a period of 24 hours. In this approach, the ACO is used to obtain the unit commitment schedule and genetic algorithm technique is used to find optimal set of parameters required for ACO. The Lagrangian multiplier method is applied to obtain the economic dispatch for the 24-hour schedule. The proposed method is tested on systems having 10 to 60 generating units to illustrate its effectiveness and simulation results are presented and compared with other methods.