A HYBRID GENETIC-ARTIFICIAL FISH SWARM ALGORITHM FOR ECONOMIC LOAD DISPATCH WITH VALVE-POINT AND MULTIPLE FUEL EFFECTS

By

Author

Presented To

Department of Engineering

ABSTRACT
The restructuring of the electrical power industry has given rise to a high degree of vibrancy and competitive market, which changed many features of the power industry. Energy resources become scarce, the cost of power generation increases, environmental concerns are raised, and an ever-increasing demand for electrical energy characterizes this now-altered scenario. In this perspective, Economic Load Dispatch (ELD) is necessitated. Strong heuristic techniques can go a long way in determining the optimum solution to such technical problems having large number of possible solutions. In the proposed research work, two heuristic algorithms namely: Genetic Algorithm (GA) and Artificial Fish Swarm Algorithm (AFSA) are hybridized to yield a more robust technique called “Hybrid Genetic-Artificial Fish Swarm Algorithm”, (HGAFSA) that is suitable for solving complex ELD problems. The technique is then applied to solve a multi-objective ELD problem involving higher order cost functions that includes the effects of valve-point loading and multiple fuel cost function. The proposed approach was validated using five standard IEEE test systems for 13, 40, 110, 140, and 160 generating unit systems. Testing of the developed HGAFSA based ELD algorithm (HGAFSAELDA) yielded reduction in fuel cost by 1.53%, 0.03%, 0.07%, 0.00012% and 1.37% for the 13, 40, 110, 140 and 160 generating units respectively. An annual savings in fuel cost of $3.254e+06, $3.8235e+05, $2135.7, $9.5563e+06, and $1.1588e+06 for the 13, 40, 110, 140, and 160-generating-units respectively were achieved over the existing best costs presented in (Pradhan et al., 2017). HGAFSA based optimization curves and the Cumulative Power Generation curves are also presented to demonstrate how the inequality constraints are satisfied by each of the generating units.