An Inversion Method for Evaluating Lightning Current Waveform Based on Time Series Neural Network

Abstract

An inversion method for evaluating lightning current waveforms from measured electromagnetic field data based on time series neural network (TSNN) is presented in this paper. The back-propagation neural network (BPNN) is also adopted to evaluate the channel-base current using measured electromagnetic field data, and comparisons of inversion results between TSNN and BPNN are presented. The inversion results are in good agreement with corresponding measured channel-base currents. The proposed method can evaluate the channel-base current in areas with complex terrain, and it is useful for studies on lightning-protection in power systems and lightning characteristics.