A novel algorithm for long-term load forecasting of distribution networks under redevelopment conditions

Multiple regression based methods are simple and popular trending methods for distribution long-term load forecasting which are employed in the cases that advanced land use data are not available. However, this method does not work properly for areas which are under redevelopment conditions. This paper proposes a novel algorithm, which can be employed in multiple regression based techniques in such conditions. This algorithm recognizes the redevelopment event in the small areas and shifts the Sshaped load growth curve to a suitable lower level using a decision making engine. The results show that employing this algorithm can significantly improve the accuracy of regression-based load forecasting methods under redevelopment conditions.