12In this work, a three-dimensional PEM fuel cell model has been developed and is used to investigate the 13 effects of water flooding on cell performance parameters. The presence of liquid water in the cathode gas diffusion 14 layer (GDL) limits the flow of reactants to the cathode catalyst layer, thereby reducing the overall reaction rate and concentration of reactants/products and cell current densities. It is determined that the gas diffusion layer 23 permeability has little or no effect on the current densities due to the diffusion dominated nature of the gas flow. 24However, through the incorporation of percolation theory based effective diffusivity model; a marked reduction in 25 the cell performance is observed which closely resembles published experimental observations. This is a reasonable 26 approximation for effects of water flooding which has been inherently used for further parametric studies. 27

Free flow channel confined by porous walls is a feature of many of the natural and industrial settings. Viscous flows adjacent to saturated porous medium occur in cross-flow and dead-end filtrations employed primarily in pharmaceutical and chemical industries for solid-liquid or gas-solid separations. Various mathematical models have been put forward to describe the conjugate flow dynamics based on theoretical grounds and experimental evidence. Despite this fact, there still exists a wide scope for extensive research in numerical solutions of these coupled models when applied to problems with industrial relevance. The present work aims towards the numerical analysis of coupled free/porous flow dynamics in the context of industrial filtration systems. The free flow dynamics has been expressed by the Stokes equations for the creeping, laminar flow regime whereas the flow behaviour in very low permeability porous media has been represented by the conventional Darcy equation. The combined free/porous fluid dynamical behaviour has been simulated using a mixed finite element formulation based on the standard Galerkin technique. A nodal replacement technique has been developed for the direct linking of Stokes and Darcy flow regimes which alleviates specification of any additional constraint at the free/porous interface. The simulated flow and pressure fields have been found for flow domains with different geometries which represent prototypes of actual industrial filtration equipment. Results have been obtained for varying values of permeability of the porous medium for generalised Newtonian fluids obeying the power law model. A series of numerical experiments has been performed in order to validate the coupled flow model. The developed model has been examined for its flexibility in dealing with complex geometrical domains and found to be generic in delivering convergent, stable and theoretically consistent results. The validity and accuracy of the simulated results has been affirmed by comparing with available experimental data.

in Wiley Online Library (wileyonlinelibrary.com).Work carried out in the last decade or so suggests that the simulators for multiphase flow in porous media should include an additional term, namely a dynamic coefficient, as a measure of the dynamic effect associated with capillary pressure. In this work, we examine the dependence of the dynamic coefficient on temperature by carrying out quasi-static and dynamic flow simulations for an immiscible perchloroethylene-water system. Simulations have been carried out using a two-phase porous media flow simulator for a range of temperatures between 20 and 80 C. Simulation domains represent 3-D cylindrical setups used by the authors for laboratory-scale investigations of dynamic effects in two-phase flow. Results are presented for two different porous domains, namely the coarse and fine sands, which are then interpreted by examining the correlations between dynamic coefficient(s) and temperature, time period(s) required for attaining irreducible water saturation, and the dynamic aqueous/nonaqueous phase saturation and capillary pressure plots. The simulations presented here maintain continuity from our previous work and address the uncertainties associated with the dependency of dynamic coefficient(s) on temperature, thereby complementing the existing database for the characterization of dynamic coefficients and subsequently enabling the users to carry out computationally economical and reliable modeling studies.

The dynamic effect in two-phase flow in porous media indicated by a dynamic coefficient τ depends on a number of factors (e.g. medium and fluid properties). Varying these parameters parametrically in mathematical models to compute τ incurs significant time and computational costs. To circumvent this issue, we present an artificial neural network (ANN)-based technique for predicting τ over a range of physical parameters of porous media and fluid that affect the flow. The data employed for training the ANN algorithm have been acquired from previous modeling studies. It is observed that ANN modeling can appropriately characterize the relationship between the changes in the media and fluid properties, thereby ensuring a reliable prediction of the dynamic coefficient as a function of water saturation. Our results indicate that a double-hidden-layer ANN network performs better in comparison to the single-hidden-layer ANN models for the majority of the performance tests carried out. While single-hidden-layer ANN models can reliably predict complex dynamic coefficients (e.g. water saturation relationships) at high water saturation content, the double-hidden-layer neural network model outperforms at low water saturation content. In all the cases, the single- and double-hidden-layer ANN models are better predictors in comparison to the regression models attempted in this work.