A Functional Approach for the Analysis of Peak Wind Loads on Houses Exposed to Hurricanes and Validation of Wind Tunnel Simulations

Virginia Polytechnic Institute And State University, Blacksburg VA

Investigators

Abstract

Field data collected under the Florida Coastal Monitoring Program (FCMP) during recent hurricane seasons on prototype buildings has revealed that measured pressure coefficients exceeded the provisions of building codes and that wind damage occurred at wind speeds well below the design wind speeds presented in these codes. Because the current code provisions for wind loads are based on the reduction of many wind tunnel observations into a few numbers, the question arises as to what constitutes a proper wind tunnel simulation of the flow in the atmospheric surface layer and associated wind loads on low-rise structures and what procedures should be followed in translating pressure coefficients as measured in these simulations to full scale applications. The objective of the proposed research is to determine whether generic wind tunnel simulations of low-rise structures can predict wind loads under hurricane conditions. This will be achieved by contrasting full scale peak pressure and wind load coefficients measured during recent hurricane seasons with model scale measurements with the same probability of non-exceedence. The comparison will be based on a probabilistic procedure that will be employed to obtain the distribution of peak pressure and load coefficients from single sample records. Different methods will be implemented and compared to determine the parameters of the probability distribution function of the peaks. The different limiting factors, including effects of turbulence intensity and differences between open and suburban terrain exposures, will also be addressed. The recent measurements conducted in Florida provide a unique database that can be used to assess the reliability of simulated wind loads. The proposed effort provides a probabilistic approach for contrasting full and model scale pressure and wind load coefficients. This approach redefines a peak wind load as measured in a single record by associating a non-exceedence probability with its value. The intellectual merit of this project is that it will enable a better comparison of field data with data from wind tunnel simulations. With respect to broader impacts, the proposed research will contribute to a more reliable prediction of wind loads and thus to the goal of reducing losses due to wind damage. The project team will work with the Institute for Business and Home Safety, a non-profit U.S. corporation established to support research and promote construction techniques for loss reduction from natural hazards. The project team will also work with the Multicultural Academic Opportunities Program (MAOP) at Virginia Tech to enable involvement by undergraduate and graduate students from underrepresented groups in this wind research.