Economic Load Dispatch in Thermal Power Plant Considering Additional Constraints Using Curve Fitting and Ann

Abstract

This paper presents a new efficient approach to economic load dispatch (ELD) problem with cost functions using curve fitting, ANN and particle swarm optimization (PSO). Economic load dispatch is one of the most important problems in power system operation. The practical ELD problems may not have fixed cost functions rather it changes with the coal quality, that make the problem of finding the global optimum difficult using any traditional mathematical approach. Therefore, curve fitting technique is used to obtain the coefficients of the cost curve. The same data is used for the training of the artificial neural network. The effectiveness of the algorithm is validated by carrying out extensive test on a power system involving 8 thermal generating units. The variation in calorific values of the coal used in different generators cause the change in coefficients of cost curve. This effect is incorporated using curve fitting, ANN and PSO approaches. The ELD problem is then optimized. The comparison shows the better results.