Investigation On Power Economic Dispatch Optimization Based On Evolutionary Computation Algorithm

In the world, the power security, reliable and economic operation play a very important role in national economic development. Economic dispatch problem of this study is one of the hard problems in the power system. The main goal of this problem is to achieve normal operation, minimizing the unit cost of electricity.In this paper, we combined the method of Evolutionary Algorithms (EAs) with economic dispatch problem and proposed several improved algorithms. The main content is as follows:We first describe the basic concept of EA and present the origin and development of these algorithms. Furthermore, we introduce PSO and Differential algorithms. We also describe the basic procedure of the two algorithms in details. At last, benchmark functions are used to compare the performance of the two algorithms.From the result, the traditional PSO shows fast convergence than other EAs. But it still has a premature problem. Then a new PSO is proposed to overcome the premature and low precision of the standard PSO. To this end, we first proposed a valued region which defines the searching direction of particles. Then the particles explore in the entire solution space with a defined probability. By testing on the economic dispatch problem with 3 and 40 points in power systems, the experiments show that the algorithm gets more precise and stable results.A new DE based on a Cauthy mutation strategy is proposed. Combing with the merit of gbest in PSO, the individuals in the new algorithm may exploit the area nearby the valuable point. Experiments show that our algorithm performs much better than the other compared algorithms not only the quality of solutions on solving the economic dispatch problem with 6 point but also more fast convergence rate.