An adaptive neural network approach in load forecasting in a power system

The authors present a method of short term load forecasting using a neural network. A three layered feedforward adaptive neural network, trained by back-propagation, is used. This method is applied to real data from a power system and comparative results with other methods are given.<
>