الفهرس 
Introduction
Holidays and special daysOnly 14 pages are availabe for public view
 |  | from | 186 |  |  |
| | | from | 186 | | |
Abstract
The main theme of this thesis is to establish an accurate and simple model for short term load forecasting. This forecast is essentially for monitoring and controlling power system operation. An adaptive model based on hourly load and weather informations is proposed. Detection for the bad data is made. The daily and weekly load correlations are onsidered. The model updating depends on a bias of the recent observations of previous forecast errors. An adaptive error estimation ( mean and variance ) is used to have the expected load on a probabilistic base.
The proposed model has the advantage of low storage requirements and quick calculations easily implemented in on-line computer system. Applications of the proposed algorithm on two another utility company in Europe are made.
Several tests are made on the proposed model for a variety of conditions including weather extremes and different times of the year. These teste show the model gives accurate results and an imorovement in accuracy W. R. T. previous models is achieved although the model is more simple.