An improved neuro-based approach for locating cloud-to-ground lightning using radiated electric field waveform data

Abstract

A new technique based on an artificial neural network (ANN) is proposed to locate cloud-to-ground lightning return stroke channels (RSCs). The technique uses a two-layer resilient backpropagation neural network to estimate the RSC-to-measuring station distance. The training of the implemented ANN is based on simulated electric field data, using the model of modified transmission line (MTL) for lightning RSC. The performance of the proposed technique is evaluated by applying it to real measured data given by Lin et al, 1979. It is shown that the technique predicts the location of RSC more accurately when the data associated with the subsequent return stroke are used. This stems from the fact the MTL model of RSC used in the training of the ANN is closer to the second return stroke than the first one