Application of PSO based ann model for STLF

This paper presents a new approach for modeling STLF (Short Term Load Forecasting) in which STLF-ANN forecaster is trained using swarm intelligence. ANN (Artificial Neural Network) has been used successfully for STLF. However, ANN-based STLF models use BP (Backward Propagation) algorithm for training which does not ensure convergence and hangs in local optima more often. Moreover, BP requires much longer time for training which makes it difficult for real-time application. In this paper, we propose smaller ANN models of STLF based on hourly load data and train it through the use of PSO (Particle Swarm Optimization) Algorithm. The approach gives better trained models capable of performing well over time varying window and results in fairly accurate forecasts. (author)