Electronic Thesis and Dissertation Repository

Thesis Format



Doctor of Philosophy


Civil and Environmental Engineering


Kopp, Gregory A.


Quasi-steady (QS) theory is a commonly-used tool in wind engineering for the assessment of wind loads on structures induced by large-scale gusts. However, the QS-based approach tends to underestimate peak loads significantly since it fails to capture the effects of small-scale and body-generated turbulence. The objective of this thesis is to develop a method to estimate peak wind loads on low-rise roofs based on a partial-turbulence approach, i.e., using the QS vector model for loads induced by large-scale fluctuations from the incident flow, and a separate statistical model to account for the effects due to body-generated turbulence.

Wind tunnel tests have been conducted for four 1:50 scaled low-rise building models with gable and hip roofs. Roof slopes range from 5:12 to 12:12. Data of similar tests on a flat roof model conducted by Wu and Kopp (2016) is taken for comparison. It is found that the smallest scale that QS vector models can reach is about 5H (H denotes the mean roof height) in length, while the largest scale affected by the body-generated turbulence can be up to 30H. The performance of QS vector models is found to be better on flat roofs than sloped roofs, and is closely related to the type of aerodynamics at different locations. Specifically, it is found to work reasonably well in regions of flow separation, but less well for flow reattachment and positive pressure zones on the windward faces of the 12:12 sloped roofs.

A statistical model has been developed to account for the pressure component induced by the body-generated turbulence, which is found to be dependent on both nominal wind directions and terrain conditions, and the differences can be minimized by normalizing it with the turbulence kinetic energy and the Quasi-Steady pressure coefficients. The final model takes the form of a 3-parameter T-scale distribution. It is valid for panels that are governed by suction loads due to flow separation and can work across roof shapes and terrains. By combining this model with the QS vector model, a method is developed to estimate peak pressure coefficients using a Monte-Carlo approach.

Summary for Lay Audience

Severe wind storms have caused significant losses in North America, because it is a major source of damage for low-rise structures. The traditional methods of estimating peak wind-induced loads on structures are generally through wind tunnel tests, which are effective for wind profiles of the atmospheric boundary layer, but meet difficulty when dealing with storms with rapidly-changing directions, for example, tornadoes and downbursts.

In this thesis, a prediction model has been developed to estimate peak wind-induced pressures on low-rise roofs, from the wind velocity data measured at an upstream location. The results indicate that this model provides reasonably good predictions for the uplift of gable and hip roofs, as well as relatively large area-averaged roof panels that subject to severe suction forces. Theoretically, this model is applicable for wind fields with rapidly-changing directions, and can be a potential solution to assess extreme loads from tornadoes and downbursts.

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License