Economic load dispatch (ED) is an important task in the power plants operation which aims to allocate power generations to match load demand at minimal possible cost while satisfying all the power units and system constraints (El-Hawary and Christensen, 1979). The complexity of the problem is due to the nonlinear and non-smooth characteristics of the input-output curves of the generators, because of valve-point effect, ramp rate limits and prohibited operating zones. The mathematical programming-based optimization methods such as lambda iteration, base point participation method, Gradient and Newton’s methods can solve successfully the convex ED problems (Wood and Wollenberg, 1996). But unfortunately, these methods are ineffective in handling the non-convex ED problems with non-differentiable characteristics due to high complexity. Dynamic programming can solve such type of problem, but it suffers from the curse of dimensionality. Hence for optimal solution, this problem needs a fast, robust and accurate solution methodology. Nowadays, heuristic search methods such as simulated annealing (SA) (El-Hawary and Christensen, 1979; and Wood and Wollenberg, 1996), genetic algorithm (GA) (Chiang, 2007), Evolutionary Programming (EP) (Sinha et al., 2003), Particle Swarm Optimization (PSO) (Zwe-Lee, 2003; Selvakumar and Thanuskodi, 2007 and 2008; and Krishna et al., 2009), Bacteria Foraging Optimization (BFO) (Ghoshal et al., 2009), Differential Evolution (DE) (Storn and Price, 1995) and Chaotic Ant Swarm Optimization (Cai et al., 2007) are employed to solve the ED problems. All the approaches have achieved success to a certain extent.
|
