DefinePK

Abstract

Forecasting is the way of knowing the future value based on some past records. In electrical power systems, there is a great need for accurate forecasting of the future load and energy requirements. Accurate load forecast provides system dispatchers with timely information to operate the system economically and reliably. It is also necessary because availability of electricity is one of the most important factors for industrial development, especially for a developing country like India. It is required to be careful that the energy forecast is neither too conservative nor too optimistic. Artificial Intelligence techniques have shown promising results in many systems. Recent progress in the applications of Artificial Neural Networks (ANN) technology to power systems in the areas of forecasting has made it possible to use this technology to overcome the limitations of the other methods used for electrical load forecasting. In this work, the GA-BPN model is used for extracting the best weight matrices for different layers of BPN thus forecasting the future power demand more accurately. For this reason, this work introduces evolution of connection weights in ANN using GA as means of improving adaptability of the forecasting.