A Tendency Based Harmony Search Algorithm And Its Application In Economic Load Dispatch Problem

Harmony Search(HS) algorithm is a kind of new meta-heuristic algorithm which improves the population and get the optimal solutions by imitating musicians who get great music by musical memory and pitch adjusting. Therefore, harmony search algorithm has strong global search capability and it has been widely used in the field of engineering optimization.Economic Load Dispatch problem(ELD) is a kind of typical optimization problem of power system planning and operation scheduling, which purpose is to minimize the cost of electricity with meeting the load and operational constraints conditions. The ELD has an important significance on the power system cost saving, improve the efficiency and reliability of power systems. However, existing optimization methods cannot obtain the satisfactory results within a reasonable time in face of non-linear, discrete, non-conductive, more constraints ELD problems.Thus, it is still an important issue to find an effective optimization method concerned by the field of power systems.This paper proposed a modified harmony search algorithm. This algorithm introduces the tendency behavior characteristics of the firefly algorithm to harmony search algorithm to improve its shortcomings, and the improved harmony search algorithm is applied to solving the economic load dispatch of power system. The main content is showed as follows:(1) Considering existing limitation of HS algorithm, this paper proposes a modified HS algorithm based on the tendency which introduces the tendency behavior of firefly algorithm. This algorithm generated new harmonies by exchanging the information between individuals within the scope of the instrument, which effectively improve the population diversity. At the same time, the stochastic process theory is used to analyze the convergence of MHS. Finally, 13 benchmark functions are used to test the performance of modified HS algorithm, and the influences of main parameters are analyzed. Compared with other improved HS algorithm, the experimental results show that the MHS has good optimization ability.(2) In order to effectively solve the economic load dispatch, this article introduces the reconstruction of the population to improve the diversity of the population in the latter part.Meanwhile, four kinds of typical economic load dispatch problem are used to verify that the performance of MHS for solving ELD problems, which include: standard ELD, ELD with valve point effect, ELD problem with prohibition zones and ELD with load constraints. Experimental results show that MHS can effectively solve these problems.