Short Term Load Forecasting Using Neural Network Trained with Genetic Algorithm & Particle Swarm Optimization

Short term load forecasting is very essential to the operation of electricity companies. It enhances the energy-efficient and reliable operation of power system. Artificial neural networks have long been proven as a very accurate non-linear mapper. ANN based STLF models generally use back propagation algorithm which does not converge optimally & requires much longer time for training, which makes it difficult for real-time application. In this paper we propose a smaller MLPNN trained by genetic algorithm & particle swarm optimization. The GA training gives better accuracy than BP training, where as it takes much longer time. But the PSO training approach converges much faster than both the BP and GA, with a slight compromise in accuracy. This looks to be very suitable for real-time implementation.