Search
Close this search box.

DC23 – External

In a few words

EOLIOS Engineering provided an overview of the thermo-aerodynamic conditions of the thermal plume of a data center and its generator set on the air coolers.

Project

Study of thermal plumes

Year

2022

Customer

NC

Location

France

Typology

Data Center

Continue browsing :

Our other projects :

Latest news:

Study of thermal plumes

This project posed several technical challenges during the design and construction phase. One wasthe release of hot air and fumes from the cooling devices and emergency generators. The relatively small space available makes it less obvious that the fresh air flow, the exhaust air and the data center fumes are always separated.

Thus, in CFD modeling the vertical profile of wind speeds is extremely important. It depends mainly on the degree of roughness of the environment, i.e. the tendency of buildings, trees and other obstacles to slow down the wind by opposing its movement. Thus over the entire thickness of the atmospheric boundary layer, the wind speed increases as one moves away from the ground. In our study, to take into account the wind shear we use a methodology similar to a physical wind tunnel by considering the urban environment for a 500m periphery around the study site and by applying a site-dependent velocity profile. This allows us to evaluate the impact of thermal plumes under different wind configurations.

Design by CFD simulation

Heat plumes in data center buildings

Identification of thermal recirculations:

The CFD simulation showed that, in this study, there was a loop between the generator set and the air coolers. Indeed, the hot gases released by the generating set are reabsorbed by the air coolers. This is problematic for the operation of these heat exchangers because the air taken in must be in the right temperature range at all times. Improvement solutions have been proposed to prevent any looping and consequently any malfunctioning.

Bouclage -surchauffe - canicules - data center -CFD
Identification of looping phenomena

Identification of hot spots

What is the origin of hot spots on data center roofs?

Numerical simulations contribute to a better understanding of thermo-aerodynamic phenomena. The major interest of CFD is to predict and verify all eventualities in order to develop relevant technical solutions. Thanks to the calculation servers that EOLIOS possesses, the models can be simulated in their entirety with great precision. This makes it possible to simultaneously analyze the impact of many phenomena on the external thermo-aerodynamic evolution of the data center. This study revealed design flaws that could be rectified.

How did CFD simulation help to optimize these systems?

Numerical simulations contribute to a better understanding of thermo-aerodynamic phenomena. The major interest of CFD is to predict and verify all eventualities in order to develop relevant technical solutions. Thanks to the calculation servers that EOLIOS possesses, the models can be simulated in their entirety with great precision. This makes it possible to simultaneously analyze the impact of many phenomena on the external thermo-aerodynamic evolution of the data center. This study revealed design flaws that could be rectified.

Temperature map of the air cooler level

Thanks to the conclusions of the study, we were able to propose to the design teams adapted solutions to solve these problems. Using a CFD simulation, you can analyze, predict, verify and correct any errors that may occur in your design. This fast and accurate method saves design time and costs. In addition, it guarantees concrete and reliable results. Integrating CFD into the design phase means bringing in experts to ensure that no problems occur in the future.

Discover other projects

Pressure loss study – Generator – Data center

Data Center – PA 22 – External

Technical premises – Data Center

Cooling optimization – Data Center

Data Centers – DC15.1 & DC15.2 – External

Data Center – Paris

Data Center – GAZ NOVEC

CFD-Data Center – Saint Denis

Data center – DC25 – Internal

Data center – DC17 – Internal

Data Center – DC10 – Internal

Data center – DC25 & DC26 – External

Data center – D14 – External

Data center – DC17 – External

DC23 – External