Numerical method for solving coupled heat and mass transfer through walls for future integration into an urban climate model
Résumé
To face up to the present energy crisis and the climate challenge, incorporating plant aggregates into reactive or clayey mineral matrices has proved to be an ecological and effective way to improve the thermal performance of buildings. It also helps to reduce both the grey energy involved and the consumption of fossil or non-renewable materials. However, the characterization and optimization of bio-based composites is relatively time-consuming, which may be a hindrance to their development. Our objective is to efficiently predict the thermal conductivity of a hemp particle as this is a fundamental input value for predicting thermal behaviour at the material scale. We found few reliable data on this topic in the literature. This is a major challenge that needs to be addressed as it can lead to erroneous results even when the prediction model is relevant and well developed. An additional difficulty arises from the fact that plant aggregate shows high variability. This paper opens new perspectives by suggesting new procedures for determining the thermal conductivity of hemp shiv using two approaches and considering, in particular, the variability of the raw material. Since existing methods for measuring thermal conductivity do not provide access to the particulate value, an experimental and a theoretical method are proposed to determine it. Both are based on analytical equations obtained by homogenization and both are shown to lead to consistent results. Comparison with the few data available in the literature also highlights a possible overestimation of the values found in previous works.