الفهرس | Only 14 pages are availabe for public view |
Abstract This thesis describes a new method for the electric load forecasting using artificial neural networks. The work has been done for both the short term and the long term load forecasting. A full study has been done on the various variables influencing the electric load. A detailed study has been done on the weather variables and their effect on individuals and the electric demand. A novel feature classification technique has been used for the analysis of the various inputs of the network. The system uses a feedforward multi-layer perceptron network that is trained using the backpropagation technique. F or the short term load forecasting the system will be used following two paradigms, one for the prediction of the weekdays and the other for the prediction of the weekends. The system produces each hour the forecast of the following twenty-four hours and uses the last load value for its forecast. The weather variables are fully implemented in the system. The system is only retrained weekly with the change from weekdays to weekend. The retraining time is well below the time limit. The adaptivity, retraining frequency and time horizon emphasizes the applicability of the system. For the long-term load forecasting, two models are used. One model to forecast the maximum demand and the other to forecast the energy consumption. The system forecasts the following ten years. The economic and demographic indices are fully implemented in the system. The data used in this work is real data obtained from a major electric utility. The data covers a large period of time. It has been used without any pre-processing. The system has been tested extensively. The tests have been done on various weather patterns in order to make sure of its ability to perform for any characteristics. The results have been analyzed not only depending on the accuracy but also on their sustainability and the system requirements to produce them. These requirements, such as data availability, time and computational requirements are realistic and simple. The use of neural networks for the electric load forecasting is shown to be superior to the conventional methods. The obtained accuracy is better than that of the conventional methods. It also does not suffer from high computational requirements or numerical instabilities. Furthermore, it combines the capability to include various variables without affecting the non-stationarity characteristic of the electric load. This work presents a system for the electric load forecasting. Its performance makes it a significant step forward towards achieving a complete and accurate knowledge about the future load. This knowledge will certainly improve the operational and planning procedures of the power system. The proposed system can be used by any utility for its load forecasting operation. It is suitable for on-line operation without the need of human experts. |