Model to Full-Scale Validation of Peak Pressure Mechanisms in Buildings that Cause Cladding Failures and Windstorm Damage

Rensselaer Polytechnic Institute, Troy NY

Investigators

Abstract

Fundamental to wind engineering is the need to understand and model the underlying physics of flow separation of wind loading on civil infrastructure (e.g., buildings), as unlike any other engineering domain, these structures are not prototyped but rely almost exclusively on wind tunnel tests at vastly reduced length and lower velocity scales. Wind tunnel results are rarely validated due to the complexity of instrumenting the structure and the long duration required to observe design events. Most windstorm damage to buildings is initiated with cladding failures at locations where very high suction pressures are observed on the building, typically near corners and edges where the flow separates from the structure under peak and fluctuating pressures. Wind tunnels have become ubiquitous for obtaining wind loads on all types of structures, but discrepancies between peak and fluctuating pressures generated in the separated flow regions on roofs of low-rise structures modeled in boundary layer wind tunnels and observed in the field have been long-reported. This project will investigate the flow mechanisms that cause these very high suctions at near full-scale using the NSF-supported Natural Hazards Engineering Research Infrastructure (NHERI) Wall of Wind (WOW) facility at Florida International University to determine if current small-scale wind tunnel tests are able to reproduce them and to determine which incoming flow characteristics are most critical in causing the high suctions. From parametric studies of separating flows, the modeling criteria critical in developing these peak pressures will be isolated and the ability to simulate them at reduced scales in typical wind tunnel studies will be ascertained. In this way, confidence in traditional wind tunnel testing can be validated and current experimental procedures improved and verified. This detailed knowledge will provide greater confidence in wind tunnel testing and generic wind tunnel data for engineers to use, and ultimately translate to reduced windstorm damage to civil infrastructure. This project will train a postdoctoral researcher; work with the Troy Middle School in Troy, New York, on a National Future Cities Competition for a "windy city;" and incorporate the research outcomes into a graduate course on wind engineering and the institution's Bedford Program on progressive buildings. This project aims to provide a mechanistic characterization of the flow physics of separating shear layers, and the effect of the full turbulence spectrum of the approach flow on peak and fluctuating pressure generation in the vicinity of separation. From a fundamental fluid mechanics viewpoint, there is a need to understand the applicability and scalability of decades of fundamental research on separated flows, which has been performed predominately at low Reynolds Numbers and in low turbulence flows, on full-scale civil infrastructure. Using the WOW facility, a large-scale bluff body will be exposed to a range of turbulent structures and intensities at very high Reynolds Numbers, and flow and pressure fields will be studied simultaneously. This project will investigate why the presence of both small-scale (high frequency) and large-scale (low frequency) turbulent structures in the freestream are needed concurrently to cause the largest magnitude pressures within the separation region and how these scale from model to full-scale. A set of systematic measurements of the unsteady characteristics of this canonical flow phenomenon at high Reynolds Numbers will be beneficial to the fluid mechanics community to understand the simulation limitations of a wide class of flow phenomenon. In addition, the results of this research will provide a benchmark for validation of computational fluid dynamics codes. With an advanced understanding of mechanics of the generation of peak and fluctuating pressures at large physical length scales and near full-scale Reynolds Numbers, guidelines and recommendations on appropriate physical modelling of the flow field will be provided; additionally, correction factors for wind tunnel results will be developed. Data from this project will be archived and made available in the NHERI Data Depot (https://www.designsafe-ci.org).