Lightning performance assessment of overhead high voltage transmission lines using an artificial neural network method and its application to the Hellenic lines

This paper presents a novel approach to the lightning performance assessment of overhead high voltage transmission lines based on artificial neural networks (ANN). A feedforward (FF) ANN method was addressed which uses the Levenberg-Marquardt learning algorithm. Actual input and output data collected from operating Hellenic high voltage transmission lines such as the tower footing resistance, the peak lightning current, the lightning current derivative, the lightning level and the number of line's lightning failures were used in the training process. The method has been applied, on several operating Hellenic transmission lines of 150 kV and 400 kV, carefully selected among others due to their high failure rates during lightning thunderstorms. The obtained results were almost identical to the field observation data collected from the Hellenic Public Power Corporation and these obtained using conventional techniques. The presented methodology can be proved valuable to the studies of electric power systems designers, intended for more effective protection of high voltage transmission lines against lightning strikes.