Parametric Reconstruction Method for the Long Time-Series Return-Stroke Current of Triggered Lightning Based on the Particle Swarm Optimization Algorithm

The lightning research group of the Chinese Academy of Meteorological Sciences has carried out observations and experiments with artificially triggered lightning for more than ten years, and it accumulated thousands of examples of channel base current data on the return stroke of artificially triggered lightning prior to 2020. Based on the current data, this paper explores ways of improving the constructed function of a long time-series (400 mu s) return stroke and the parametric reconstruction of the current waveform. The long time-series current can be divided into three components: the breakdown pulse current, corona current and quasi-uniform current. The corona current and quasi-uniform current are constructed by a Heidler function, while the breakdown pulse current is determined by the waveform characteristics of the current peak, which can be divided into the Heidler type and high-order exponential type. According to the three-component model of the return-stroke current, a two-step parametric reconstruction method for the long time-series lightning return-stroke current based on the particle swarm optimization (PSO) algorithm is proposed. In this paper, the results of the parameterized reconstruction of 14 return strokes are given, and the results of the multidimensional error analysis of 13 long time-series return strokes are given to illustrate the accuracy of the improved parameterized model of the return-stroke current and the effectiveness of the two-step reconstruction method based on the PSO algorithm.