In this paper, two probabilistic analyses are performed to try to gain a better understanding of the damage inflicted on galvanized roofs from hail strikes. The first analysis addresses the probability that a hailstone will have the energy required to cause yielding in steel sheets. The second probability study lays the theoretical groundwork for determining the probability of a loss in structural integrity of the steel roof. Data and equations are combined from various sources to create a separate limit state function for each analysis. The probabilistic analyses are performed in NESSUS. For determining probability of yielding, an energy related method for determining stress in an impacted plate is used in conjunction with LS-DYNA finite element software to discover that all perpendicular hail strikes will induce some degree of plastic deformation on simply supported steel sheets in the gauge range of galvanized steel roofs. More data is still required for successful determination of probability of a loss in structural integrity, but the foundation is established, which would allow probabilistic determination of the most economic roof gauges in a given region. There appears to be a logarithmic relationship between the analytically determined initial impact force of a hailstone and the indentation diameter that remains after striking in which the thickness of the sheet plays an important role, but more data points need to be collected to verify this.
Probabilistic Analysis of Steel Roof Damage From Hail Strike
- Views Icon Views
- Share Icon Share
- Search Site
Shimek, E, Ekwaro-Osire, S, & Rasty, J. "Probabilistic Analysis of Steel Roof Damage From Hail Strike." Proceedings of the ASME 2011 International Mechanical Engineering Congress and Exposition. Volume 9: Transportation Systems; Safety Engineering, Risk Analysis and Reliability Methods; Applied Stochastic Optimization, Uncertainty and Probability. Denver, Colorado, USA. November 11–17, 2011. pp. 503-510. ASME. https://doi.org/10.1115/IMECE2011-64640
Download citation file: