Forecasting energy consumption using particle swarm optimization and gravitational search algorithm

摘要

A hybrid algorithm for short-term load forecasting is proposed. The particle swarm optimization algorithm used in the training phase of the artificial neural network is optimized by combining it with the gravitational search algorithm. In this paper, we have combined the exploitation of PSO and exploration of GSA to form a single algorithm that can be used to get more accurate results for load forecast. The results reflect that hybrid algorithm avoids local minimum and have better convergence speed than the PSO algorithm and GSA algorithm individually.