Short-Term Load Forecasting Based on GA-PSO Optimized Extreme Learning Machine

In order to improve accuracy of short-term load forecasting and solve low precision of the traditional single load-forecasting model and the shortcoming of common intelligent algorithms are easy to fall into local minimization in solving high-dimensional complex problems. In this paper, a short-term load-forecasting model based on the GA-PSO optimized extreme learning machine(ELM)is established. ELM connection weight and the hidden layer threshold are optimized with GA-PSO algorithm to avoid local optimal solution due to random initialization during the learning process, thus enhancing ELM forecasting performance. Finally, the practical example shows effectiveness of the proposed model.