GPU-Based Lightning Attachment and Detachment Points Prediction for Aircraft

• Published in: 2024 IEEE International Conference on Computational Electromagnetics (ICCEM) pp. 1 - 3
• Main Authors: Zhang, Ping, Wang, Ming, Gong, Linshu, Wei, Zeyong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 15.04.2024
• Subjects: Adaptation models; Atmospheric modeling; Computational modeling; Graphics processing units; Lightning protection; Poisson equations; Predictive models
• ISSN: 2768-2552
• DOI: 10.1109/ICCEM60619.2024.10559156

According to regulations, airliners are required to possess lightning protection capabilities. The attachment and detachment points of lightning on an airliner are the primary factors influencing the distribution of lightning currents on the airliner's surface, and they are crucial elements that must be clarified in lightning protection design. The location of attachment points is typically determined through lightning zone tests, which are costly and time-consuming. Traditional simulation methods, such as the finite difference method (FDM) for solving the Poisson equation of the electrostatic field combined with the Monte Carlo method, can accurately predict the location of lightning attachment and detachment points. However, the computational speed is not ideal for large-scale and highly detailed models of airliners. This paper proposes a lightning attachment and detachment points prediction model based on GPU acceleration, which can be applied to predict attachment and detachment points on complex airliner models, achieving an efficiency improvement of over 100 times. The research results contribute to the establishment of a low-cost, rapid, and flexible lightning protection design capability.