Dhari Y. Al-Samaraee; Sammar J. Ismail
Abstract
As far as electrical power system is concerned, there has been a need to find out the future load in advance. Load forecasting has been a central integral process, throughout planning ...
Read More ...
As far as electrical power system is concerned, there has been a need to find out the future load in advance. Load forecasting has been a central integral process, throughout planning and operation of electrical utilities. An approach of artificial neural networks (multi layer perceptron) to short term load forecasting is presented in this work. Four different architectures of neural networks have been trained and tested to forecast the daily peak load of Baghdad city. A historical daily peak load and weather data were proposed for the forecasting process. A back propagation algorithm has been used to train these networks. MATLAB version 6.1 program was used.