سال انتشار: ۱۳۸۷

محل انتشار: بیست و سومین کنفرانس بین المللی برق

تعداد صفحات: ۷

نویسنده(ها):

N Ghaffarzadeh – Electrical Engineering Department, Iran University of Science and Technology, Tehran
S Jamali – Electrical Engineering Department, Iran University of Science and Technology, Tehran

چکیده:

Short term electric load forecasting is necessary for optimum operation planning of power generation facilities, as it affects both system reliability and fuel consumption. Short term load forecasting involves forecasting load demand in a short term time frame. The short time frame may consist of half hourly prediction up to weekly prediction. In this paper, a short term load forecasting realized by wavelet networks is proposed. It can predict the hourly load accurately. As a case study, the Pennsylvanian hourly load data is used for training of the wavelet network. The effectiveness of this method has been tested using practical daily load data. The proposed approach is compared to multi layer perceptron neural networks with Back Propagation training algorithm. The simulation results show that the presented intelligent technique for load forecasting cangive satisfactory results