Investigation of Lightning Attachment Risks to Small Structures Associated with the Electrogeometric Model (EGM)

William Brooks, David H. Barnett, W. A. Harrison, David Hattz, John Mankowski, James Dickens, Andreas Neuber

Research output: Contribution to journalArticlepeer-review

Abstract

A 3-D Monte Carlo-type random walk model was constructed for the assessment of lightning attachment probabilities to small structures. The simulation assumed buildings had a negligible impact on the propagation of lightning. A purely stochastic propagation model based on a previously proposed gas and charged particle diffusion process was employed. The attachment was based on the electrogeometric model in which striking distance is determined by return stroke peak current. This model allows for hundreds of thousands of samples to be evaluated in the window of a few minutes on readily available consumer computing hardware. Using this model, it became possible to enable characterizing building protection as a probability distribution of striking distance. Such was done to provide a deeper understanding of the impact of building protection design choices than is readily available from binary testing. The model was calibrated for minimum input resolution, which is found to be insensitive to variations in step length and moderately insensitive to variations in propagation angle distribution, resulting in normalized errors of less than 15% (rms). A parametric sweep of geometric features was performed for a large (100 m × 50 m) rectangular building with catenary wire protection. For heights of less than 30 m, lightning was found to bypass protection structures and strike to the building itself at rates that were insensitive to variations in building height. The extent to which the protection may be recessed from the building's perimeter was found to have a significant impact. Variations in building aspect ratio were found to be of limited impact except for cases of extreme aspect ratio where competition with the ground appears to have resulted in much better lightning protection performance.

Original languageEnglish
Article number9089302
Pages (from-to)2163-2174
Number of pages12
JournalIEEE Transactions on Plasma Science
Volume48
Issue number6
DOIs
StatePublished - Jun 2020

Keywords

  • Buildings
  • computer-aided analysis
  • electrical breakdown
  • lightning
  • simulation
  • stochastic approximation

Fingerprint

Dive into the research topics of 'Investigation of Lightning Attachment Risks to Small Structures Associated with the Electrogeometric Model (EGM)'. Together they form a unique fingerprint.

Cite this