ライブラリ登録: Guest

ISSN Online: 2377-424X

ISBN Print: 978-1-56700-421-2

International Heat Transfer Conference 15
August, 10-15, 2014, Kyoto, Japan

Direct-Inverse Analysis of Heat Transfer in Micro-Models of Porous Media

Get access (open in a dialog) DOI: 10.1615/IHTC15.pmd.009730
pages 6871-6882

要約

Micro-models of porous media have been recently constructed and employed in the proposition and analysis of conceptual models for contaminated soils remediation and enhanced oil recovery from mature reservoirs. Different micro-fabrication techniques have been utilized in constructing such pore scale two- and threedimensional physical models of the porous medium to be studied, also known as ROC (rock-on-a-chip), followed by the metrological and hydraulic characterization of the micro-model. The present work employs a combination of integral transforms and Bayesian inference in the proposition of a direct-inverse problem analysis for the characterization of two-dimensional micro-models of porous media, based on the availability of non-intrusive measurements of the external surface wall temperatures. The physical dimensions and properties are taken from an actually constructed micro-model, manufactured through micro-milling on a PMMA substrate, according to tomography images of an actual rock sample, and sealed with a thermal pressing machine. An infrared camera is assumed to be employed to capture the temperature distributions on the external surface of the microfluidic chip, while flowing hot water through the micro-model. A conjugated heat transfer model is proposed for the conduction-convection analysis which provides the direct problem solution for the theoretical external surface temperatures, which is then combined to the inverse analysis through the MCMC approach with the Metropolis-Hastings algorithm, so as to identify the hydraulic and thermal parameters of the tested micro-model. The direct-inverse problem with the non-intrusive temperature measurements is then analyzed for different porous media configurations of known geometries and thus different dimensionless parameters, so as to demonstrate the consistency of the proposed approach in the parameters identification and characterization of the micro-porous model.