Proceedings of the Second International Energy 2030 Conference,
November 4-5, 2008, Abu Dhabi, UAE

Electric Energy Forecasting by WEB-Based Method

Balanthi Beig
The Petroleum Institute, UAE

Majid Poshtan
The Petroleum Institute, UAE

Rajesh Ramanand
The Petroleum Institute, UAE

Abstract
The paper presents a web-based system to forecast the demanded Electric Energy in power systems. The first method used was the linear regression method to find the slope and y-intercept, and then uses those values to predict the load for the next time interval. The second method was a back propagation neural network. This method learns patterns from the existing data set, and the trained network is then used to predict future values of the load. Overall, this program is a practical program for load forecasting for planning and operating of power systems. It has also the advantage of an easy access to the system load from the Internet for operating centers. This feature will give an open access to the future load for load marketing in deregulated systems. It may also help the interconnected utilities to predict the demanded load from each other with a reasonable error.