Evaluation of the reliability of a heat and mass transfer model in hygroscopic material. (October 2019)

• Record Type: Journal Article
• Authors: Berger, Julien; Busser, Thomas; Reddy, Sohail; Dulikravich, George S.
• Is Part Of: International journal of heat and mass transfer. Volume 142(2019)
• Publication Date: 2019-10
• Subjects: Heat and mass transfer -- Porous media -- Bench-marking with experimental data -- Du Fort–Frankel scheme -- Parameter estimation problem; Heat -- Transmission -- Periodicals; Mass transfer -- Periodicals; Chaleur -- Transmission -- Périodiques; Transfert de masse -- Périodiques; Electronic journals; 621.4022
• DOI: 10.1016/j.ijheatmasstransfer.2019.06.014
• Languages: English
• ISSNs: 0017-9310

Abstract

Highlights: Reliability of model for heat and mass transfer in porous material is achieved. An efficient numerical model is proposed to save computational efforts. Inverse problem is solved to estimate uncertain input parameter. Identifiability of the seven unknown coefficients is demonstrated. Other experimental data are used to benchmark the numerical prediction of the model. Abstract: The reliability of a model is its accuracy in predicting the physical phenomena. In this paper, the robustness of a model of heat and mass transfer in a porous material is evaluated by comparing the numerical predictions with experimental observations. An experimental facility composed of an enclosure made with spruce CLT panels is used. An increase of temperature is applied in the inside air volume to force the heat transfer from the inner to the outer surfaces. Sensors inside the material enables to have experimental observations of the physical phenomena. Before bench-marking the numerical model, a first set of experimental data is used to reduce the two major source of uncertainties in the model. Indeed, the first source arises from surface heat and mass transfer coefficients, usually determined by empirical correlations. The second comes the thermal conductivity of the material which is defined through standard methods as invariant for the three layers of the spruce panels. To overcome this issue, a set of seven uncertain parameters are estimated using an hybrid optimizer after demonstrating their theoretical and practical identifiability. Then, the reliability of the numerical model, based on the Du Fort–Frankel explicit scheme, is evaluated by comparison to a second set of experimental data obtained in another wall of the enclosure. A very satisfactory agreement is remarked showing the accuracy of the model to predict the physical phenomena in this hygroscopic porous material.