A New Controllable Weak Recycling Inflow Turbulence Generator for Evaluating Wind Effects on Building in LES
Jinghan Wang,Chao Li,Lingwei Chen,Shengtao Zhou,Gang Hu,Jinping Ou
DOI: https://doi.org/10.1016/j.engstruct.2024.118742
IF: 5.582
2024-01-01
Engineering Structures
Abstract:The weak recycling method, known as the "rescaling-recycling" method, has been extended to simulate microscale turbulent atmospheric boundary layers (ABL) by employing staggered blocks in the developing section, replicating physical ABL wind tunnel setups. However, this method faces challenges, such as the need to reevaluate the sizes and configurations of roughness elements when addressing different objective structures, computational domain sizes, or ground exposures. To address these issues, we propose a controllable weak recycling (CWR) method for generating turbulent ABL flows over rough surfaces in large eddy simulation (LES). This method can directly produce a turbulent boundary layer with arbitrary target characteristics in a single simulation, significantly enhancing generation efficiency and reducing computational resource consumption. Our refinement method incorporates a feedback proportional control algorithm into the "rescaling" procedure to minimize errors between simulated and targeted turbulence intensities. Besides, an existing drag model is utilized in the near-ground region to maintain horizontal homogeneous ABL flows over rough surfaces. The proposed CWR method is validated by simulating three turbulent ABL flows over various rough terrains using LES. Results confirm that the generated inflow turbulence adheres to prescribed mean velocities and turbulence intensities, demonstrating consistent self-sustainability along the flow direction. Furthermore, the generated wind spectra closely align with the anisotropic von-Karman spectrum. When applying the extracted wind time histories at the main domain's inlet, we find that the mean and fluctuating wind pressures on a high-rise building closely match specified wind tunnel test results.