Multi-gradient PSO algorithm for economic dispatch of thermal generating units in smart grid

We propose a novel algorithm called, multi-gradient particle swarm optimization (MG-PSO), for solving economic dispatch (ED) problem of thermal generating units (TGUs) under smart power grid constraints. The curve of cost function of TGUs becomes non-convex when these are subjected to ramp rate limits and prohibited operating zones. The proposed MG-PSO algorithm is able to solve such complex problem. In MG-PSO algorithm, different negative gradients are used. These negative gradients are used as guides for m particles in the search of global minima. The diversity in negative gradients is a key of the MG-PSO algorithm. Due to this diversity, the m particles cover largest search area as much as possible. The velocity vectors of the m particles are significantly affected by only one negative gradient called, the best negative gradient among all used negative gradients. This makes the m particles adjust their positions and improve their direction according to the best negative gradient. The performance of the MG-PSO algorithm has been verified on 6 and 15 TGUs test systems. The proposed MG-PSO algorithm gives good quality and promising results in solving the ED problem. In addition, the MG-PSO algorithm produces better results in terms of fitness values when compared with PSO algorithm and other optimization techniques.