Hybrid krill herd optimizer for thermal power scheduling problem

A hybridized meta-heuristic technique is applied to solve Economic-Environmental Power Dispatch (EEPD) problem. Krill Herd Algorithm (KHA) is a meta-heuristic technique of swarm intelligence based on populations of krill individuals and its motion for searching food. To improve the convergence characteristics of KHA, it is combined with a confined selective operator, termed as the Hybrid Krill Herd Optimizer (HKHO). In this technique, the krill's position is upgraded with confined krill individuals instead of the arbitrarily chosen individuals as processed in basic KHA. This proposed HKHO technique prevents entrapping of best possible solution in confined optima which means, it avoids the premature convergence of optimal solutions. A non-interactive multi-objective optimization technique is applied whereby the price penalty factor is applied to get scalar objective optimization in case of EEPD problem. The HKHO is implemented in small and medium standard test systems to show the applicability to solve EEPD problem. The developed optimizer is applied to validate the results on two power systems consisted of 6- and 40- thermal units. It gives 2.27% savings in fuel cost and 13.3 % reduction in emission of pollutants for 6-thermal units' power systems with respects to the results undertaken for comparison. Whereas, 40-units' power system, depicts the conflicting nature of the objectives, when the fuel cost is decreased by 0.16% and emission of pollutants decreases by 0.04%. In both the cases, the achieved results are comparable to already published work in terms of fuel cost and emission of pollutants as shown in tables of comparative analysis of achieved results. The examination of the results shows the satisfactory improvement in best possible solution.