Particle swarm optimization with time varying acceleration coefficients for non-convex economic power dispatch

Economic dispatch (ED) is one of the key functions of the modern energy management system. Conventional gradient based methods can solve the ED problem effectively only if the fuel cost curves of generating units are assumed to be piecewise linear, monotonically increasing in nature, otherwise these methods are likely to converge to suboptimal or infeasible solutions. Classical particle swarm optimization (PSO) algorithm is capable of achieving near global solutions for such problems but it tends to converge prematurely. The practical NCED problem is solved here using PSO with a novel parameter automation strategy in which time varying acceleration coefficients (TVAC) are employed to efficiently control the local and global search, such that premature convergence is avoided and global solutions are achieved. The performance of this method has been compared and found to be superior compared to the results of a few PSO variants and some recently published results.