الفهرس | Only 14 pages are availabe for public view |
Abstract Load forecasting problem is receiving great and growing attention as being an important and primary tool in power system planning and operation. Importance of load forecasting becomes more significant in developing countries with high growth rate such as Egypt. The accuracy of load forecasting is crucial due to its direct influence on generation planning, and for its economical impacts. The objective of this thesis is to perform both long and short term load forecasting based on real historical data for the Egyptian unified network. To get these forecasts we used one of the artificial intelligence techniques which is the artificial neural network. On the other hand, it will be compared to one of the traditional methods which is the regression model. Performance of both models will be investigated including effect of weather factors and the results will be compared to obtain the validation of the proposed techniques. The software used for designing an operation of ANN is MATLAB 7.5 .The Egyptian load curve will be carefully analyzed to obtain useful data. Many experiments will be done by changing neural network parameters and the results will be observed. Different inputs will be tested using statistical analysis using SPSS software (Statistical Package for Social Sciences). A comprehensive discussion will be held about load affecting factors in Egypt at different time frames. For regression model both univariate and multi-variate models will be used. |