Hybrid Harmony Search and Modified Harmony Search Optimizations Performance in Economical Load Dispatch

Abstract

New optimization methods are an essential tools to analyze the performance of modern hybrid grids appropriately. Multiple hybridizations are performed in the basic conventional optimization methods available such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Differential Evolutionary (DE), and Harmony Search (HS). Limitations noticed in existing methods, (i) high complexity and large convergence time in GA, (ii) large convergence time in case of PSO, (iii) initial population size for DE, and (iv) frequent use of memory in HS. The Modified Harmony Search (MHS) algorithm reaches a global optimum solution on harmony memory, and Hybrid Harmony Search (HHS) is of PSO, DE, and HS. Both methods are applied in the Economical Load Dispatch requirement (ELD). Initially, proposed methods, PSO, and GA have been applied for a six-unit system, which includes six thermal units, forty-six transmission paths, and twenty-six buses. It is observed that the proposed MHS provides a better loss minimization. Further study has been performed on fifteen, twenty, and forty-unit systems using the MHS method. It is observed that in the fifteen-unit system, the proposed MHS method provides improved performance in terms of both loss and cost minimization 42as compared to PSO [1] and GA (RANKSUM method). In twenty and forty-unit systems, MHS gives a better solution.