Short Term Load Forecasting Based on IGSA-ELM Algorithm

This paper proposes a novel short-term load forecasting (STLF) method based on extreme learning machine (ELM) and improved gravitational search algorithm (IGSA). The IGSA is used to search the optimal set of input weights and hidden biases for the ELM, improving the basic gravitational search algorithm (GSA) by involving the ability of exploitation in particle swarm optimization (PSO). Based on the IGSA-ELM algorithm, a model to predict the maximal load data of the next day has been established. Meanwhile, comparative experiments between the IGSA-ELM and some conventional methods have been carried out. The results show that the proposed method performs better than the existing algorithms in terms of speed and accuracy.