الفهرس Only 14 pages are availabe for public view
 |  | from | 97 |  |  |
| | | from | 97 | | |
Abstract
forecasting is very essential to the operation of electricity companies. It enhances the energy-efficient and reliable operation of a power system. Load forecasting plays a key role in helping an electric utility to make important decisions on power flow, load switching, voltage control, network reconfiguration, and infrastructure development.
This thesis presents a study of short-term load forecasting. The short-term load forecasting is of importance in the daily operations of a power utility. It is required for unit commitment, energy transfer scheduling, load flow and load dispatch. The development of an accurate, fast and robust short-term load forecasting methodology is of importance to both the electric utility and its customers.
A supervised artificial neural network (ANN) has been used in this work.
ANN has been replacing traditional methods in many applications offering, besides a better performance, a handful of advantages: needs no system model, tolerates bizarre patterns, notable adaptive capability and so on.
The Supervise Control and Data Acquisition (SCADA) provides historical load data obtained from Alexandria regional control center for the months of July and August from 2003 to 2006 to be used in the forecasting. The main advantages of SCADA are the pre-processing of the data sets, and the accuracy in collecting data. The inputs used for the neural network are the previous hour load, previous day load, previous week load.
The goal of this thesis is to develop two forecasting models for the hourly and daily electric load demand in Alexandria using the data extracted from SCADA and the forecaster XL software. Forecaster XL software is a neural network forecasting add-in for MS Excel. It is designed especially for easy and reliable forecasting by automatic data preprocessing and neural network preparation. These models can be used by any dispatcher/operator in the control center any time without need neither experience nor knowledge in neural network. The main contribution of the thesis is to provide a simple, accurate, and applicable methodology for load forecasting which is essential for proper load flow analysis.