Artificial Neural Network Based Short Term Load Forecasting for the Distribution Utilities

The load forecasting is a tool of utmost important for the power industry as it can influence areas like power generation and trading, infrastructure development planning etc. Implementation of the load forecasting tool in the distribution utilities has a wider impact up to the power generation level. The load forecasting has been an area in power systems where the human experts are still performing better than the algorithms which have been put forward as alternatives. Many techniques have been put forward for the accurate load forecasting. Different Artificial Neural Networks (ANN) with different architectures have been proposed in the last few years for load forecasting purpose resulting in a large number of publications on this subject. In this paper a new Feed Forward Neural Network has been proposed for the short term load forecasting in correlation with the variations in weather, for distribution management systems. The proposed neural network can forecast the load profile with a lead time of one to seven days.