Abstract:Lightning is an important cause of casualties, and of the interruption of power supply and distribution facilities. Monitoring lightning locations is essential in disaster prevention and mitigation. Although there are many ways to obtain lightning information, there are still substantial problems in intelligent lightning monitoring. Deep learning combined with weather radar data and land attribute data can lay the foundation for future monitoring of lightning locations. Therefore, based on the residual network, the Lightning Monitoring Residual Network (LM-ResNet) is proposed in this paper to monitor lightning location. Furthermore, comparisons with GoogLeNet and DenseNet were also conducted to evaluate the proposed model. The results show that the LM-ResNet model has significant potential in monitoring lightning locations. In this study, we converted the lightning monitoring problem into a binary classification problem and then obtained weather radar product data (including the plan position indicator (PPI), composite reflectance (CR), echo top (ET), vertical integral liquid water (VIL), and average radial velocity (V)) and land attribute data (including aspect, slope, land use, and NDVI) to establish a lightning feature dataset. During model training, the focal loss function was adopted as a loss function to address the constructed imbalanced lightning feature dataset. Moreover, we conducted stepwise sensitivity analysis and single factor sensitivity analysis. The results of stepwise sensitivity analysis show that the best performance can be achieved using all the data, followed by the combination of PPI, CR, ET, and VIL. The single factor sensitivity analysis results show that the ET radar product data are very important for the monitoring of lightning locations, and the NDVI land attribute data also make significant contributions.

Lightning Location and Peak Current Estimation From Lightning-Induced Voltages on Transmission Lines With a Machine Learning Approach

Machine Learning-Based Lightning Localization Algorithm Using Lightning-Induced Voltages on Transmission Lines

A Regionalized Evaluation Method for Indirect Lightning Stroke Flashover Rate in 10kv Distribution Networks

Estimating flashover occurrence in distribution lines: A novel approach focused on the current-peak-to-distance ratio

Lightning Performance Evaluation of Transmission Line Based on Data-Driven Lightning Identification, Tracking, and Analysis

Precise Lightning Strike Detection in Overhead Lines Using KL-VMD and PE-SGMD Innovations

A Lightning Fault Location Method for Transmission Lines Based on By-Base Tower Law

Novel Distributed Lightning Fault Location Method for Transmission Line Based on Non-contact Measurement Device of Lightning Current

A Lightning Current Measurement Method Based on Optical Sensing Technology

Identification of Direct Lightning Strike Faults Based on Mahalanobis Distance and S-transform

A New Tool for Lightning Performance Assessment of Overhead Transmission Lines

Lightning risk assessment method of transmission line using distributed traveling wave monitoring

An Analytical Method for Estimation of Lightning Performance of Transmission Lines Based on a Leader Progression Model

A Generic Severity Estimation Method for Lightning-caused Voltage Sags by Using Multiplatform Monitoring Data Mining

Monitoring Lightning Location Based on Deep Learning Combined with Multisource Spatial Data

Analysis and research prospect of lightning-induced voltages in overhead transmission lines

Lightning Strike Risk Classification of Power Distribution Line Tower Based on Support Vector Machine

Lightning Fault Location of 10kV Distribution Network With Multistage Branch Lines

A Novel Method to Locate a Fault of Transmission Lines by Shielding Failure

Classification of VLF/LF Lightning Signals Using Sensors and Deep Learning Methods

Research on lighting fault identification technology of transmission lines based on non-contact sensing