In this article we describe
how CFD was used in order to optimize the thermal comfort
of a residential building. The objective of the study was
to determine whether the arrangement of airconditioning
and heating devices was sufficient to keep the house comfortably
cool in the summer and warm in the winter. By ensuring
that all devices are properly sized and located, it is
possible to minimize the energy costs of operating the
system, thereby minimizing both the environmental and economic
impact of the development. The methodology described delivers
useful design information and is both simple and inexpensive
to apply. Pro S3 is a company located in Turin, Italy providing
support to industrial companies using advanced design for
product innovation.
The building in question is largely open-plan and has an irregular layout.
The large living area is split by a mezzanine floor and is joined to the
kitchen area without doors. The thermal environment of the resulting large
volume occupied zone is difficult to control. The design of the building
disrupts the usual insulating near the roof air layer so that the volume
is subject to large solar gains through the roof during the summer and large
heat loses through the roof during the winter. Economic and environmental
constraints restrict the number of airconditioning units to two and the number
of radiators to three. A simple 3D CAD model was constructed from the 2D
drawings of the building.
Representative, low detail, furniture is included in the model in order to
determine how much influence it has on the predicted thermal environment.
The computational mesh was constructed automatically from the CAD model using
trimmed cell technology. Air conditioning units are simulated using coupled
pairs of inlet and outlet boundary conditions, controlled via user coding.
The mass flow and the temperature of the air entering the room are calculated
from the air-conditioning unit data sheet. The radiators are simulated using
baffle cells with an exchange area that is equivalent to those of the real
heaters.
Appropriate levels of thermal resistance are applied to the windows, walls
and roof. Adiabatic boundary conditions are prescribed at all internal walls.
The analysis
Several calculations were performed. The initial simulations were used to
tune the modeling parameters to optimize the convergence behavior of subsequent
calculations. In order to accurately model the effects of buoyancy, a transient
solution method was employed, using a pressure correction under-relaxation
factor of 0.8, double precision and a small time step. Temperatures were
monitored at the head height of a typical occupant in a number of locations.
Simulations were performed for both winter and summer scenarios.
The summer conditioning results
One of the major aims of these analyses is to verify if the airconditioning
unit placed in the mezzanine floor of the living room can adequately cool
the whole room. The analysis clearly shows that if this unit is placed strategically,
it can contribute dramatically to the comfort of the occupants of the lower
living room level. The results also demonstrated, however, that the area
beneath the mezzanine balcony is exposed to relatively stale air, when compared
with other areas of the building.
The winter heating results
The results from the winter heating scenarios were easier to predict. There
is a great difference in temperature between the lower level and the mezzanine
floor. The mezzanine level is heated excessively by the large buoyant plume
generated by the heater in the open area away from the balcony.
Conclusions
These analyses clearly show that, even in a relatively simple building geometry
such as this, it is possible to predict interesting thermal effects that
might not have otherwise been apparent until the house was built. A practical
methodology has been defined that is reusable for other similar cases. With
parametric CAD and STAR-CD's automatic meshing tools, it is possible to generate
a good computational mesh in a small amount of time. Recalculating the simulations
with air-conditioning units and radiators in different positions takes just
a few hours. The complete analysis with animated gifs can be seen on our
web site: www.pros3.it
|
