Numerical analysis of the most appropriate heat transfer correlations for free ventilated double skin photovoltaic façades
ELSEVIER Applied Thermal Engineering Vol. 57 Issues 1-2 Pag. 57-68 August 2013
a CIMNE, Building, Energy and Environment Group, 08224 Terrassa
b Barcelona Supercomputing Center (BSC-CNS), 08034 Barcelona
c Applied Physics Section of the Higher Polytechnic School (EPS), University of Lleida, Jaume II 69, 25001 Lleida
Double skin façades with photovoltaic integrated systems are building components which combine functions of the building envelope with electricity and thermal energy generation. The heat transfer modelling of these components, especially under free convection situations, raises a high complexity and is one of the main drawbacks for a massive dissemination of this technology. Many attempts to fill this gap have been undertaken and some mathematical correlations allowing evaluating average Nusselt numbers and air mass flow rate have been obtained in the last decades. However, very few studies faced a detailed analysis of the valid range of these mathematical expressions and of the restrictions entailed.
This paper introduces a methodology to analyse the valid range of the existing mathematical correlations for the convective heat transfer coefficients and for the air mass flow rate in laminar and transition to turbulent free convection, and provides an evaluation of the effect of the asymmetry of the wall boundary conditions. A specific numerical code, based on a stabilized finite element formulation (FEM), is used to solve the incompressible Navier–Stokes equations within the air gap and to determine the accuracy of the existing heat transfer correlations. This evaluation was preceded by an extensive bibliographic research as well as a detailed validation of the physical and numerical hypothesis adopted in the finite element code.
- This paper focuses on the heat transfer modelling of double skin PV façades.
- The validity range of existing heat transfer mathematical correlations is analysed.
- A numerical code is used and calibration with experimental measures is carried out.
- Asymmetrical boundary conditions affect heat transfer only for small Rayleigh numbers.
- Asymmetrical boundary conditions affect air mass flow rate for high Rayleigh numbers.