CH-18-C014 -- Design Optimization of Active Chilled Beam for an Office Space using Large Eddy Simulation

Use of active chilled beams for cooling and heating applications has drawn significant interests in North America in recent years due to its benefits such as easier integration with ceiling, lower air-flow, less ductwork and lower cost. The optimal performance of active chilled beams is dependent on many factors such as its placement, primary supply air-flow, mixed-air throw angle, and effective induced air mechanism. The design optimization of an active chilled beam is difficult to achieve with conventional one-dimensional calculations, and performing experimental tests for many design variables is cost-prohibitive. Computational fluid dynamics (CFD) offers cost-effective and accurate design optimization tool for active chilled beam systems. One critical limitation of using CFD is employing empirical turbulence modeling approach like one-equation or two-equation Reynolds-averaged Navier-Stokes (RANS). Large eddy simulation (LES), with proper flow-resolving computational grids, can provide more accurate prediction of flow distribution in such scenarios. In the present work, high fidelity large eddy simulations are carried out to analyze active chilled beam design for a typical office space application. The computational modeling approach is first validated with available experimental data. The limitation of two-equation RANS modeling approach is illustrated by comparing results with LES and published experimental data. The CFD model includes buoyancy, and typical heat loads for manikins, laptops, lights etc. Several CFD modeling scenarios are created and simulations performed for understanding the impact of assumptions on results. Finally, flow structures are analyzed to assess the cooling performance in built environment using active chilled beams, based on metrics like thermal comfort and draft risk. The induction ratio (IR) was accurately predicted by LES while RANS predicted IR within 8% error, when compared to experimental data. LES model captured the detailed flow structures and jet width relatively better. Results also confirmed draftless cooling of the office space.