Due to the increasing windage area of container ships, wind loads are playing a more important role in navigating the ship at open sea and especially through harbor areas. This paper presents 3D steady RANS CFD simulations of wind loads on a container ship, validation with wind-tunnel measurements and an analysis of the impact of geometrical simplifications. For the validation, CFD simulations are performed in a narrow computational domain resembling the cross-section of the wind tunnel. Blockage effects caused by the domain boundaries are studied by comparing CFD results in the wind tunnel domain and a larger domain. The average absolute difference in numerically simulated and measured total wind load on the ship ranges from 37.9% for a simple box-shaped representation of the ship to only 5.9% for the most detailed model. Modeling the spaces in-between containers on the deck shows a 10.4% average decrease in total wind load on the ship. Modeling a more slender ship hull while keeping the projected front and side area of the ship similar, yields an average decrease in total wind load of 5.9%. Blockage correction following the approach of the Engineering Sciences Date Unit underestimates the maximum lateral wind load up to 17.5%.
The CFD simulation of droplet evaporation in turbulent flows is challenging as the accuracy and reliability of the results strongly depend on the available sub-models and their modeling parameters. This study presents a systematic sensitivity analysis focused on the impact of the most widely used discrete random walk (DRW) model, the constant time scale coefficient (CL), the turbulence model, and the drag coefficient model. CFD simulations with the Eulerian-Lagrangian approach are employed. The analysis is based on grid-sensitivity analysis and validation with measurements of spray evaporation in a heated turbulent airflow. The results show that the use of the DRW model leads to a good agreement between the CFD results and the experimental data of droplet size and droplet mean velocity, attributed to the turbulent fluctuations inducing droplet dispersion. The best performance is observed for the standard k-ε turbulence model with CL = 0.30 and 0.45. It is mainly attributed to the reasonable interaction time between droplets and turbulent eddies at these two CL values. The three drag coefficient models (i.e., Spherical, Ischii-Zuber and Grace) lead to similar results due to the low droplet Reynolds number.
In both experimental and numerical studies of urban air quality, idealized sources are commonly employed to reproduce the traffic emissions, and green infrastructures (GI) are widely used as an approach for pollution mitigation. However, previous studies have found that (i) idealized sources may not be adequate to reproduce the pollutant emitted from realistic car sources (RCS), and (ii) GI may not necessarily have a positive impact on air quality in street canyons. The goal of this study is to investigate the impact of different GI configurations on pollutant concentrations in an urban street canyon with RCS by computational fluid dynamics (CFD) simulations. First, CFD results obtained from the scale-adaptive simulation (SAS) in terms of mean velocity and concentration are validated with wind-tunnel (WT) data for idealized line sources with vegetation. Next, SAS simulations are performed on a full-scale street canyon with RCS to evaluate the effectiveness of different GI configurations in reducing pollutant concentration. These GI configurations include a different number and arrangement of hedgerows, also combined with trees deployed in the center of the canyon. The SAS results indicate that the number of hedgerows (from one to three) can reduce pollutant concentration by 1.8% to 13.4% compared to a street canyon without GI. Hedgerows are generally more effective in reducing the pollutant concentration at the pedestrian level of a street canyon than combinations of trees and hedges. This study can support urban planners in enhancing urban green spaces and reducing pollutant levels in urban environments.
Heat and vapour convective surface coefficients are required in practically all heat and vapour transport calculations. In building envelope research, such coefficients are often assumed constants for a set of conditions. Heat transfer surface coefficients are often determined using empirical correlations based on measurements of different geometry and flows. Vapour transfer surface coefficients have been measured for some specific conditions, but more often, they are determined with the Chilton-Colburn analogy using known heat transfer coefficients. This analogy breaks down when radiation and sources of heat and moisture are included. Experiments have reported differences up to 300%. In this paper, the heat transfer process in the boundary layer is examined using Computational Fluid Dynamics (CFD) for laminar and turbulent air flows. The feasibility and accuracy of using CFD to calculate convective heat transfer coefficients is examined. A grid sensitivity analysis is performed for the CFD solutions, and Richardson Extrapolation is used to determine the grid independent solutions for the convective heat transfer coefficients. The coefficients are validated using empirical, semi-empirical and/or analytical solutions. CFD is found to be an accurate method of predicting heat transfer for the cases studied in this paper. For the laminar forced convection simulations the convective heat transfer coefficients differed from analytical values by ±0.5%. Results for the turbulent forced convection cases had good agreement with universal law-of-the-wall theory and with correlations from literature. Wall functions used to describe boundary layer heat transfer for the turbulent cases are found to be inaccurate for thermally developing regions.
Drafting is riding close behind each other to reduce aerodynamic drag. New simulations and measurements for drafting cyclists show that also the leading cyclist experiences a drag reduction, up to 3.1%. For six or more similarly-sized drafting cyclists, the position enjoying the largest drag reduction is the one-but-last position.
Computational Fluid Dynamics (CFD) is increasingly used for natural ventilation studies because it provides whole-flow-field data, allows full control of the boundary conditions, and does not suffer from similarity constraints. In addition, it allows efficient parametric studies and the simultaneous (i.e. coupled) simulation of outdoor wind flow and indoor air flow. In this paper, CFD for coupled wind-induced natural ventilation is evaluated by comparison with detailed wind tunnel measurements with Particle Image Velocimetry (PIV). The focus is a parametric analysis of the influence of the inflow turbulent kinetic energy and the near-wall treatment on the predicted indoor mean velocity pattern. The 3D steady Reynolds-averaged Navier-Stokes (RANS) equations are solved with the RNG k-e turbulence model for a generic isolated building model with large openings. It is shown that the influence of the inlet turbulent kinetic energy on the mean velocity is very large (up to a factor 5), while the influence of the near-wall treatment is limited to at most 130%.
