Focusing on careful air distribution, temperature uniformity and pollutant control inside enclosed environments, mixing ventilation is a fundamental engineering method for indoor environmental control. Within the field of developing computational techniques, a recent benchmark paper “Benchmark Tests for a Computer Simulated Person [1]” presents a novel approach for assessing simulated entity performance via VALIDATION Study. This work investigates the complex measures of computational simulation using ANSYS Fluent with an eye on the subtle powers of computer-generated human representations. The importance of the work is found in its methodical methodology to evaluate across challenging scenarios simulation accuracy, responsiveness, and adaptive potential. Establishing strict testing procedures helps the benchmark study to link theoretical computational models with real-world interaction scenarios, therefore giving researchers and engineers a strong tool for evaluating digital human simulation technologies.

• Reference [1]: Nielsen, Peter V., et al. “Benchmark tests for a computer simulated person.” Aalborg University, Indoor Environmental Engineering(2003).

Figure 1: Four manikins for experiments.

Simulation Process

ANSYS Fluent was used for the Computational Fluid Dynamics (CFD) simulation, which included a full geometric model of a seated CSP in a wind tunnel configuration (2.44m × 2.46m × 1.2m). The computational domain was discretized with a focus on boundary layer refinement around the CSP, which is critical for accurately capturing thermal plume and flow interactions. The mesh technique used inflated layers near the CSP surface to resolve the viscous sublayer, resulting in acceptable y+ values for wall treatment. In Case 2, the input velocity was set at 0.20 m/s with a temperature of 22°C. Turbulence parameters were defined as ko = 9.6E-03 m²/s² and εo = 3.1E-04 J/kg·s. The CSP was set up with a total heat flow of 76W (equivalent to 1 MET activity level), with 38W dedicated to convective heat transfer. This arrangement is consistent with the benchmark standards for assessing thermal comfort and flow characteristics in mixed ventilation scenarios.

Figure 2: Plane outline at z = 0 of the tunnel and location of the CSP. The CSP is seated in the mixing ventilation case and facing the flow [2]

Post-processing

The numerical simulation shows excellent agreement with experimental benchmark data, particularly in the velocity magnitude profile comparison. The CFD findings (orange line) closely match the experimental data (black squares) across the vertical measurement locations, with a maximum difference of less than 10%. There is a strong link between peak velocities at z ≈ 0.6m and z ≈ 0.75m, as well as the velocity minimum at z = 0.15m. This validation demonstrates the model’s capacity to effectively anticipate flow patterns in mixed ventilation conditions, particularly around the Computer Simulated Person (CSP).

Figure 3: Benchmark validation plot – Mixing Ventilation CFD Simulation by ANSYS Fluent

Flow visualization illustrates difficult aerodynamic interactions surrounding the CSP, including separate recirculation zones and thermal plume effects. The streamline patterns show two principal vortex formations: one over the CSP’s head and one in the lower cervical region, with speeds ranging from 0 to 7.9 m/s. The cross-sectional view focuses on flow separation and reattachment locations around the CSP geometry, where the entering uniform flow (0.20 m/s) undergoes significant alteration due to body interference and temperature effects. These flow structures are consistent with expected mixing ventilation features and highlight the model’s ability to capture both bulk flow and local thermal plume dynamics.

Figure 4: Velocity streams in Mixing Ventilation CFD Simulation by ANSYS Fluent