Dual analysis for heat exchange: Application to thermal bridges - Université de technologie de Compiègne Access content directly
Journal Articles Computers & Mathematics with Applications Year : 2018

Dual analysis for heat exchange: Application to thermal bridges

Samer Taoum
  • Function : Author
Emmanuel Lefrançois
  • Function : Author
  • PersonId : 867462

Abstract

The research presented in this article is a contribution to the characterization of thermal dual analysis. The aim of a dual analysis based on two complementary numerical methods is to calculate the stationary temperature field. The two methods are respectively the classical Kinetically Admissible (KA) finite element or temperature method, and the Statically Admissible (SA) flow method. The originality of this work is twofold. First, it uses a flow approach (SA) in 2D and 3D that naturally respects flow conservation (in contrast to the KA method). Second, calculating the dissipation energy allows the exact solution of a problem to be enclosed, with the lower and upper bounds being calculated respectively by the KA and SA methods. Thermal bridges are areas of risk in the building walls, because they can give rise to uncontrolled increments in heat transfer. A number of studies have looked at thermal bridges, but much of the numerical code for building energy simulations uses heat transfer models based on one-dimensional heat flow analysis. Fabrizio Ascione (2014) showed that this can lead to unreliable results. The originality of this paper is to consider 2D and 3D thermal bridges using dual analysis with mesh sizes smaller than those used in conventional finite element approaches.
No file

Dates and versions

hal-01993636 , version 1 (25-01-2019)

Identifiers

Cite

Samer Taoum, Emmanuel Lefrançois. Dual analysis for heat exchange: Application to thermal bridges. Computers & Mathematics with Applications, 2018, 75 (10), pp.3471-3487. ⟨10.1016/j.camwa.2018.02.011⟩. ⟨hal-01993636⟩
37 View
0 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More