HYDRO POWER PLANT WATER INFLOW FORECASTING USING NEURAL NETWORKS

In the paper, a new method for forecast of natural inflow into the reservoir of the upstream hydro power plant is described. Water inflow forecasting is usually based on the precipitation data collected by the ombrometers situated in the river basin. Due to highly non-linear nature of mathematical relation between the amount of precipitation and water inflow, the problem to be solved is rather complex. In the paper, a new approach to water inflow forecasting based on neural networks is presented. First, selection of input parameters is discussed. Next, preparation of needed data is described and the most appropriate architecture of the neural network is chosen. Finally, efficacy of the proposed method is tested for a practical case and some results are presented. For that purpose, preliminary investigation on advantages of implementing the neural network based algorithm for water inflow forecast was conducted for Soca river hydro system in Slovenia.