Short Term Load Forecasting for Electrical Dispatcher of Baghdad City Based on SVM-PSO Method

• Published in: 2018 2nd International Conference on Electrical Engineering and Informatics (ICon EEI) pp. 140 - 143
• Main Authors: Jaber, Aqeel S., Satar, Koay A., Shalash, Nadheer A.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-10-2018
• Subjects: Forecasting; Lighting; Load forecasting; Particle swarm optimization; Power system stability; PSO; Support vector machines; SVM
• DOI: 10.1109/ICon-EEI.2018.8784316

The short-term of power forecasting represents as one of the substantial roles in the safe and economic uses of the power system. The climate consideration is the main challenge which is facing the improvement of the load forecasting accuracy. In this paper hybrid PSO and SVM methods used to forecast non-linear load data was proposed. First, description of the detail of a hybrid method between PSO and support vector machines learning method. Secondly; applying this method for forecasting the load in Bagdad city. Finally, the proposed method implemented by MATLAB and compared with classical support vector machines method. The results show that the proposed method was very clear in the accuracy of the forecasting depends on terms of absolute proportional error (MAPE).