اضغط هنا لتنزيل الملف

<!--

<!--<!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:Arial; mso-bidi-theme-font:minor-bidi;} </style> <![endif]-->

An  intelligent information system based on neural networks for sales forecasting

Ashour F. Nosier

Faculty of Computers and Information Systems, Libya

E-Mail: [email protected]

ABSTRACT

One of the major problems facing the electric utility is the unknown future demand of energy sales that needs to be forecasted correctly. A forecast analysis is important since it is used to make investment decisions today concerning future developments. The uncertainties associated with the future make this process a difficult task, especially over long term horizon. The accuracy of the forecasting models depends on the functional relationship between energy sales and the factors affecting on it. This paper presents an Intelligent Information System (IIS) for energy sales forecasting. The IIS contains historical database, user interface, statistical modules, and Artificial Neural Network (ANN) module. The historical database contains historical data for energy sales, no of population, economic indices and peak loads. The IIS can be used for single-step and multi-step ahead forecasting. A comparative analysis between three of the most common conventional forecasting models and ANN system is performed. Results of annual sales forecasting for Libya energy system are presented to demonstrate the effectiveness of IIS. The comparison shows that ANN system outperforms the statistical techniques. The developed IIS is considered as a prototype for energy system planning managers to support them in the process of determining the expected annual sales figures.

Key words: Intelligent Information System, Feed-forward Neural Network, Energy Sales Forecasting, and Statistical Forecasting Models.