Study on Multi-Station Identification Technology of Lightning Electromagnetic Pulses (LEMPs) Based on Deep Learning

Given the increasing threat of lightning to modern electronic systems and human activities, the accurate identification and classification of lightning electromagnetic pulses has become a critical research focus, prompting the present study. A dataset was established by collecting lightning electromagnetic signals generated by various types of lightning under diverse environmental conditions via the lightning location system of the Institute of Electrical Engineering, Chinese Academy of Sciences. Subsequently, A deep learning model integrating a convolutional neural network was developed for feature extraction and pattern recognition using the multi-station data. Experimental results demonstrate that the proposed model significantly improves LEMP identification accuracy (exceeding 97%) compared to existing single-station methods. Moreover, it effectively uncovers complex hidden features within the data, outperforming conventional approaches in both accuracy and robustness. In conclusion, the proposed deep learning model offers a reliable technical foundation for lightning monitoring and localization based on LEMP signals.