STOCHASTIC SIMULATION OF CORRELATED WIND PRESSURE FIELDS ON LOW-RISE GABLE ROOF STRUCTURES

Abstract eng:
This paper presents the preliminary results of an ongoing study to quantify the effects of wind loads on the roof of a low-rise building. Wind tunnel data is used to establish a statistical characterization of the pressure coefficients sampled at 14 points in a region of separated flow over a single sheathing panel. The overall fluctuating uplift on the sheathing panel is a function of the statistical properties at the individual taps on that panel. Of particular interest for design is the most likely peak value of the aggregate load in extreme wind conditions, which is dependent on the probability density function at each tap as well as the degree of correlation between taps. A simulation methodology is applied to generate realizations of wind loads that duplicate the measured characteristics over the sheathing panel. Both the cross-spectral and highly non-Gaussian characteristics of the data at 14 pressure taps are maintained through the use of a recently developed multi-variate non-Gaussian simulation algorithm. The level of correlation between taps is varied in the simulations to investigate the effects of assuming cases from no correlation to full correlation on the panel. The ASCE 7-98 wind loads are calculated for comparison with these generated aggregate loads. Experimentally determined maximum uplift capacities of sheathing panels are also compared with the simulated loads. Two different wind directions are considered. One direction leads to Gaussian loads over the panel, and another produces non-Gaussian loads that results in increased peak loads.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket