An Effective Approach to ANN-Based Short-Term Load Forecasting Model Using Hybrid Algorithm GA-PSO

In recent years, research of optimal algorithms for short-term load forecasting has become popular. Optimal algorithms can improve forecast results in two main directions: precision and speed of convergence. To achieve both of these goals, researchers often use a combination of different algorithms. This paper describes a new combination of two optimal algorithms: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The forecasting of the 24-hour daily-load on the normal working day of the SPC-Southern Power Corporation (a big local Vietnam Company) showed a significant improvement in the error as well as the speed of convergence with several selected days. In the most accurate case, the average error of prediction reaches 1.15% while the biggest error is 3.74% and the smallest is 0.02%