A Short-Term Load Forecasting Model Based on LS-SVM Optimized by Dynamic Inertia Weight Particle Swarm Optimization Algorithm

  • Authors:

  • Dongxiao Niu;Bingen Kou;Yunyun Zhang;Zhihong Gu

  • Affiliations:

  • College of Business and Administration, North China Electric Power University, Beijing, China 102206;College of Business and Administration, North China Electric Power University, Beijing, China 102206;Department of Economics and Management, North China Electric Power University, Email:410513582@qq.com, Baoding, China 071003;College of Business and Administration, North China Electric Power University, Beijing, China 102206

  • Venue:
  • ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part II
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Short-term load forecasting is very important for power system, how to improve the accuracy of load forecasting is the keystone people pay attention to. A combined model of least squares support vector machines optimized by an improved particle swarm. Optimization algorithm is proposed in this paper to do the short-term load forecasting. Least squares support vector machines (LS-SVM) are new kinds of support vector machines (SVM) which regress faster than the standard SVM, they are adopt to do the forecasting here, and an improved particle swarm optimization (PSO) algorithm is employed to optimize the parameters gamma and sigma of LS-SVM, the new PSO outperforms the standard PSO especially in the search of orientation because of the dynamic inertia weight. A real case is experimented with to test the performance of the model, the result shows that the proposed algorithm can reduce training error and testing error of LS-SVM model, so to improve the accuracy of load forecasting.