Effect of internal structure on dynamically coupled heat and moisture transfer in closed-cell thermal insulation

Abstract

Closed-cell thermal insulation is a common insulation material with low thermal conductivity. When temperature and humidity gradients are present, moisture penetration, condensation, and accumulation might occur in the material, thereby affecting the system's thermal performance. In traditional methods used for analyzing coupled heat and moisture transfer, it is difficult to quantify the effective transmission coefficient and isolate the impact of the related structures. A multi-scale model to predict the effect of the internal structure of closed-cell thermal insulation (phenolic) on the process of dynamic heat and moisture transfer is presented in this paper. Based on a reconstructed geometry that describes the internal structure of closed-cell thermal insulation, the relationships between the multiple structure parameters and effective transmission coefficients are analyzed using the lattice Boltzmann method. From the simulation data, a database is derived. A neural network algorithm is then used to predict the effective transmission coefficients at the mesoscopic scale. Furthermore, a multi-scale model was developed by combining the mesoscopic effective transmission coefficients with the macroscopic dynamically coupled heat and moisture model. The effects of the internal structure on the process of dynamic heat and moisture transfer are analyzed using the multi-scale model. Based on the simulation results, the optimal structures of the phenolic thermal insulation for minimum heat loss and low moisture ingress are selected and discussed.