Artificial neural networks application for current rating of overhead lines

This paper describes an application of an intelligent system consisting of an expert system and artificial neural networks (ANN) for the evaluation of the thermal rating and temperature rise of overhead power lines. The hourly solar irradiance is determined by the ANN and regression best-fitting techniques. The neural network was trained for the prediction of hourly or instantaneous values of the irradiance dependent on astronomic and meteor-climatic conditions. The developed intelligent system can be used to assist operators in loading of power transmission lines in different operating, ambient, cloud and ground reflection conditions. It can also assist the operators to determine the permissible duration of the conductor overload