Multi-Objective Dispatch of Thermal System using Dragonfly Algorithm

Two-third of electricity is produced from thermal power plants and indeed major sources of environment pollution in electricity generation sector. Serious concern towards environmental protection has turned economic dispatch to economic dispatch optimization. In economic emission dispatch, both objectives i.e economy and emission reduction are satisfied at the same time. Multi-objective is the complex optimization task of achieving generation schedule with economy and reduced emissions. This paper presents the solution of multi-objective dispatch problem of thermal system using Dragonfly Algorithm. Dragonfly Algorithm is a new population based meta-heuristic algorithm mimic the swarm behavior of dragonflies. The algorithm is tested in 10-unit system on MATLAB 2010b platform. In this paper, weighted sum method has been used to get the best combination of optimum values of both objectives. Valve point effect has been considered to reflect the realities of thermal units. In the first case simple economic load dispatch has been solved on 10 unit system with Dragonfly Algorithm and results are compared with Differential Evolution algorithm. In second case Dragonfly Algorithm was implemented to solve non-convex multi-objective dispatch problem for 10 unit system and the simulation results are compared with other algorithm including DE, MODE, PDE and found effective in achieving global optimal solution. The demand is set to 2000MW for both the cases. A result shows that Dragonfly Algorithm has good convergence for calculating global optimal solution and avoided local minima problem.