Ventilation/cooling of a power generation engine hall


SimTec constructed a Computational Fluid Dynamics (CFD) model for the simulation of the temperature distribution inside a power generation hall in the Libyan desert operating with three diesel engines. The model was requested by ”LDK Consulting Engineers and Planners” (LDK). The aim of the simulation was the verification of the ventilation/cooling system designed by LDK, in the form of specific maximum temperature requirements on the diesel engines’ surfaces.

Available 2D CAD files (side and plan views) were used for the reconstruction of the 3D geometric model of the engine hall, containing only the significant, with respect to fluid flow and heat transfer, geometric details (Fig. 1). The computational mesh that was created contained 900k cells, mostly tet-dominated due to the complexity of the modeled mechanical equipment. The mesh had a higher density near the walls, particularly at surfaces of the diesel engines, in order to resolve the thermal boundary layer

The physical model provided for the turbulent air flow with buoyant heat transfer. The heat dissipated from the equipment (mostly the diesel engines) was modeled as surface heat source, evenly distributed on the respective surfaces. Solar radiation was also added to the model, according to the worst weather scenario for this geographic location.

LDK proposed a non-conventional cooling system layout (Fig. 2), comprising of two duct systems: (a) supply ducts directly over each engine and (b) three separate duct lines at a distance from the engines for the cooling of the rest of the hall space. The air flow rates and temperature supplied from the Air Handling Units were implemented as boundary conditions. The cool air, after picking up the heat from the floor-mounted equipment exits the hall from large vents at the roof of the building.

Four different simulations were performed, with different engines in operation: (a) all three engines, (b) only engine 1, (c) only engine 2 and (d) both engines 1 and 2. Case (a) proved, as expected, to be the worst in terms of high air temperatures. For this case, Fig. 3 shows the airflow streamlines injected from the Lower Line ducts (Fig. 2), colored with the value of time after injection, whereas Fig. 4 presents the isosurface of 59 [oC] temperature in the hall space. A cross-sectional temperature field is shown in Fig. 5 and the isosurface of the age of air of 300 [s] is given in Fig. 6 (the air is considered to be of zero age when injected from the supply grilles into the hall space and attains its maximum age when exiting the hall from the outlet vents).