Numerical modeling and simulation of PEM fuel cells: Progress and perspective

This paper provides a comprehensive review on the research and development in multi-scale numerical modeling and simulation of PEM fuel cells. An overview of recent progress in PEM fuel cell modeling has been provided. Fundamental transport phenomena in PEM fuel cells and the corresponding mathematical formulation of macroscale models are analyzed. Various important issues in PEM fuel cell modeling and simulation are examined in detail, including fluid flow and species transport, electron and proton transport, heat transfer and thermal management, liquid water transport and water management, transient response behaviors, and cold-start processes. Key areas for further improvements have also been discussed.     

Experimental and numerical analysis of spray-atomization characteristics of biodiesel fuel in various fuel and ambient temperatures conditions

The purpose of this work is to reveal the effects of fuel temperatures and ambient gas conditions on the spray-atomization behavior of soybean oil methyl ester (SME) fuel. The spray-atomization behavior was analyzed through spray parameters such as the axial distance from the nozzle tip, local and overall Sauter mean diameter (SMD). These parameters were obtained from a spray visualization system and a droplet measuring system. In addition, the experimental results were compared with the numerical results calculated by the KIVA-3V code. It was revealed that the increase of the fuel temperature (from 300 K to 360 K) little affects the spray liquid tip penetration. The increase of the ambient gas temperature (from 300 K to 450 K) caused a increase in the spray liquid tip penetration. Also, biodiesel fuel evaporation actively occurred due to the increase in the fuel temperature and the ambient gas temperature. Of special significance was that the highest vapor fuel mass concentration was observed at the center region of the spray axis. In the results of the microscopic characteristics, the detected local droplet size at the axial direction and overall droplet size at the axial and radial direction in a control volume increased when the fuel temperature increased. This is believed to be due to an increase in the number of small droplets that quickly evaporated. In addition, the increased fuel temperature caused the decrease of the number of droplets and the increase of the vapor fuel mass. The mean axial velocity of droplets decreased with increasing fuel temperature.     

Numerical simulation of effects of flow maldistribution on heat and mass transfer in a PEM fuel cell stack

A 3D numerical study was carried out to analyze flow, heat and mass transfer first in a single half-cell cathode channel of proton exchange membrane (PEM) fuel cell. From practical point of view, it is necessary to put the appropriate number of cells in a stack. Hence, the above study on a single half-cell is extended to a stack of channels. Due to stacking, the assumption of uniform flow distribution would no longer hold true. Therefore, the channel flow-maldistribution is considered. The water formed at the active surface due to the electrochemical reaction diffuses through the porous layer and eventually enters the gas flow duct. The higher gas velocities in the duct result in faster water vapour removal which leads to a lower value of water vapour into the duct and hence a lower Nusselt number.     

Experimental and numerical studies on the burning of aluminum micro and nanoparticle clouds in air

An experimental study has been conducted to determine flame propagation velocities in clouds of micro- (4.8 μm) and nano- (187 nm) aluminum particles in air at various concentrations. The experimental results show faster flame propagation in nanoparticle cloud with respect to the case of microparticles. Maximum flame temperature has been measured using a high-resolution spectrometer operating in the visible range. Analysis of combustion residual shows that nanoparticles combustion is realized via the gas-phase mechanism. A three-stage particle combustion model has been proposed based on these observations. Model parameters have been fitted to match the experimental results on the flame velocity and maximum temperature. Particle burning time is estimated from the flame simulations.     

Bubble dynamics and interactions with a pair of micro pillars in tandem

This study investigates flow patterns and bubble dynamics of two-phase flow around two 100 μm diameter circular pillars in tandem, which were entrenched inside a horizontal micro channel. Bubble velocity, trajectory, size, and void fraction were measured using a high speed camera and analyzed using a particle tracking velocimetry method. A range of gas and liquid superficial velocities were tested, resulting in different bubbly flow patterns, which were consistent with previous studies. These flow patterns were altered as they interacted with the pillars. Depending on the relative transverse location of bubbles to the pillars, and through bubble–bubble interaction, the flow sometimes returned to its original state. It was also determined that the pillars altered both the bubble trajectory and void fraction, especially in the pillars region.     

Experimental,numerical and analytical studies of abrasive wear: correlation between wear mechanisms and friction coefficient

The transport of granular material often generates severe damage. Understanding the correlation between the friction coefficient, particle geometry and wear mechanisms is of primary importance for materials undergoing abrasive wear. The aim of this study is to investigate the effect of particle geometry on wear mechanisms and the friction coefficient. Numerical and analytical simulations and experimental results have been compared. The process to be studied is the scratch made by a rigid cone with different attack angles on a 5xxx aluminium alloy (Al–Mg) flat surface. A scratch test was used and the wear mechanisms were observed for different attack angles. A numerical study with a finite element code was made in order to understand the effect of attack angle on the friction coefficient. The contact surface and the friction coefficient were also studied, and the results compared to the Bowden and Tabor model. The superposition of the numerical, analytical and experimental results showed a better correlation between the wear mechanisms and the friction coefficient. It also showed the importance of the model hypothesis used to simulate the scratch phenomenon. To cite this article: S. Mezlini et al., C. R. Mecanique 333 (2005).     

Effect of placing different obstacles in flow fields on performance of a PEM fuel cell: numerical investigation and experimental comparison

In this study a complete two-dimensional model for proton exchange membrane (PEM) fuel cells was used to investigate the effect of using different obstacles on the performances, current density and gas concentration for different aspect ratios (ARs). The proposed model is a full cell model, which includes all the parts of the PEM fuel cell, flow channels, gas diffusion electrodes, catalyst layers and the membrane. Also a series of tests are carried out to investigate and validate the numerical results of the polarization curve under the normal conditions. A PEM fuel cell with 25 cm² active area and Nafion 117 membrane with 4 mg Pt/cm² for the anode and cathode is employed as a membrane electrode assembly. The results show that the predicted polarization curves by using this model are in good agreement with the experimental results. Also the results show that the local current density reduces more obviously at a higher overpotential than at a lower overpotential because of the more obvious reflection phenomena in the downstream region. At lower operating voltage conditions, the overall cell performance decreases as the AR decreases.     

Theoretical analysis and numerical simulation for the spill procedure of liquid fuel of fuel air explosive with shell

Addressing the problem of spill of liquid fuel of fuel air explosive (FAE) with shell, this paper deduces variation relationship of fracture radius and fracture velocity based on continuity and motion equations of shell using the Taylor fracture criteria of cylindrical shells, and analyzes the correctness of the approximating process in detail. Furthermore, analytical solutions and numerical simulations for the liquid fuel spill process were obtained using Adomian's decomposition method (ADM) and the Runge-Kutta numerical method. Results show that both the numerical solution and the analytical solution based on ADM are closer to the result from experiments and computations from commercial dynamics numerical software.     

Experimental and numerical studies of creasing of paperboard

A laboratory creasing device to capture the most important properties of a commercial rotary creasing tool was designed. Finite element analysis of the creasing of a multiply paperboard in the laboratory crease device was presented. The multiply paperboard was modeled as a multilayered structure with cohesive softening interface model connecting the paperboard plies. The paperboard plies were modeled by an anisotropic elastic–plastic material model. The purpose of the analysis of the laboratory creasing device was to present material models that represent paperboard, and to investigate how well the analysis captured the multiply paperboard behavior during laboratory creasing. And to increase the understanding of what multiply paperboard properties that influence the laboratory crease operation. The result of the simulations showed very good correlations with the experimental obtained results. The results indicated that the paperboard properties that have the most influence is the out-of-plane shear, out-of-plane compression and the friction between the laboratory creasing device and the paperboard.     

无氧铜层裂的实验与数值研究

利用一级气体炮对国产的无氧铜（OFHC）进行了平板撞击致层裂实验,3 mm厚的OFHC飞片撞击6 mm厚的OFHC靶,采用锰铜应力计记录试样/有机玻璃（PMMA）界面附近的应力历史,获得试样发生层裂的信息。软回收试样,观测回收试样的层裂片。采用基于空穴聚集的层裂模型,数值模拟这些平板撞击致层裂实验。数值模拟的应力剖面以及试样层裂片厚度与实验结果基本一致。此外,对于国外相关的OFHC层裂实验,采用基于空穴聚集的层裂模型也作了相应的数值模拟,并进行了比较。     

Experimental and numerical studies of underwater shock wave attenuation

The attenuation of an underwater shock wave by a thin porous layer is studied both experimentally and numerically. The shock waves are generated by exploding 10 mg silver azide pellets and the pressures at different distances from the explosion center are measured. Measurements are also carried out with a gauze layer placed between the explosion source and the pressure gauge. The results with and without the gauze layer are compared evaluating the shock wave attenuation. Numerical simulations of the phenomenon are also carried out for a simple wave attenuation model. The results are compared with the experimental data. Despite the simple mathematical model of wave attenuation, the agreement between the experimental and numerical results is reasonable.Received: 22 October 2002, Accepted: 17 June 2003, Published online: 5 August 2003PACS: 47.11.+j, 47.40.Nm, 47.55.Mh     

Experimental and numerical investigation of aerodynamics of a pneumatic nozzle for suspension fuel

1. Download : Download high-res image (277KB)
2. Download : Download full-size image

     

Lattice Boltzmann modeling of transport phenomena in fuel cells and flow batteries

Fuel cells and flow batteries are promising technologies to address climate change and air pollution problems. An understanding of the complex multiscale and multiphysics transport phenomena occurring in these electrochemical systems requires powerful numerical tools. Over the past decades, the lattice Boltzmann(LB) method has attracted broad interest in the computational fluid dynamics and the numerical heat transfer communities, primarily due to its kinetic nature making it appropriate for modeling complex multiphase transport phenomena. More importantly, the LB method fits well with parallel computing due to its locality feature, which is required for large-scale engineering applications. In this article, we review the LB method for gas–liquid two-phase flows, coupled fluid flow and mass transport in porous media, and particulate flows. Examples of applications are provided in fuel cells and flow batteries. Further developments of the LB method are also outlined.     

Experimental and numerical studies of shock wave focusing in water

The focusing of plane shock waves and spherical blast waves in water by parabolic and ellipsoidal reflectors is studied experimentally and numerically. The numerical computations are performed by a second order method based on an extension of Godunov's method, which is called piecewise-linear method. Tests of this method and a comparison with pressure measurements show a strong dependence of the maximum pressure attained in the focal region on the mesh size. Outside this focal region, however, good agreement between the pressure measurements and the result of the numerical computation is found. Furthermore it is shown by numerical computation, that non-linear effects in the focusing process may be partly avoided by an optimized reflector contour which depends on the strength of the incident shock wave. This modification provides that the maximum pressure is attained at the geometric focus and not at a location between reflector surface and geometrical focus.     

Experimental and numerical studies of impact on filament-wound composite cylinder

This study focused on the impact behavior of carbon-fiber-wrapped composite cylinders subjected to impact from flat-ended,hemispherical-nosed and conical-nosed impactors.Damage morphologies of the cylinders and mechanisms of the damage were analyzed.Change laws of the maximum impact forces,durations of impact processes and energies absorbed by the cylinders after impact with different impactors and impact energies were obtained.A finite element model was developed and the simulation results were in reasonable agreement with the tests.Finally,taking the flat-ended impactor as an example,stress distributions of the cylinders under pressurization and impact were discussed.     

Experimental and numerical studies of weak blast waves in air

Effects of viscosity and vibrational nonequilibrium on the profile of a weak, spherical N-wave in air are experimentally and numerically studied. Weak blast waves were generated, in a quiescent air dome, by spark discharges and exploding wires and observed by high frequency response microphones over 40 meters. Some similarity relationships were obtained from the blast wave experiments. For observed N-waves having less than 100 Pa peak overpressure, the peak overpressure p _f and the duration of the positive phaset _d+ are found to vary with the radial distance from the sourcer as p _f r ^–1.38 andt _d + r ^0.19, whilst the rise time of the blast wave t _f linearly increases with distance. Similar trends were also found for the negative phase of the blast wave. Numerical simulations were carried out to compare with the blast wave data. The Navier-Stokes equations for spherical symmetric flows were solved by coupling with a relaxation equation for vibrational excitation of oxygen using the random choice method (RCM) adapted to supercomputing with an operator splitting technique. The resultant N-wave profiles are in good agreement with the experimental results. The numerical results clearly indicate that the wave-easing process due to the dispersive effect of vibrational relaxation plays a dominant role in determining the rise time of the N-wave.This article was processed using Springer-Verlag T_EX Shock Waves macro package 1990.     

Experimental and numerical analysis of two-phase infiltration in a partially saturated soil

The purpose of this study is to analyze the effects of the soil air flow on the process of water infiltration in a 93.5 cm deep vertical column for varied boundary conditions at the surface - positive time constant head; time constant fluxes smaller and greater than saturated soil hydraulic conductivity.Several experiments conducted on a sandy soil column with and without a possible air flow through the wall are presented. Continuous and simultaneous measurements of water content and air and water pressure heads at different depths allow the analysis of the air and water movements within the soil and the determination of the capillary pressure and relative permeability for each phase as functions of the volumetric water content.A numerical solution of the equations describing the simultaneous flow of air and water is compared with the experimental data and with the traditional one-phase flow modeling. The results show that the air movement may significantly affect water flow variables such as infiltration rates, water content profiles, and ponding times.Furthermore, some basic assumptions used in two-phase flow modeling, such as the hydrodynamic stability of the wetting fronts and the pertinence of the relative permeability concept, are discussed in the light of the experimental data.     

Experimental and numerical investigations of pressure drop in a rectangular duct with modified power law fluids

Numerical solutions are presented for fully developed laminar flow for a modified power law fluid (MPL) in a rectangular duct. The solutions are applicable to pseudoplastic fluids over a wide shear rate range from Newtonian behavior at low shear rates, through a transition region, to power law behavior at higher shear rates. The analysis identified a dimensionless shear rate parameter which, for a given set of operating conditions, specifies where in the shear rate range a particular system is operating, i.e. in the Newtonian, transition, or power law regions. The numerical results of the friction factor times Reynolds number for the Newtonian and power law region are compared with previously published results showing agreement within 0.05% in the Newtonian region, and 0.9% and 5.1% in the power law region. Rheological flow curves were measured for three CMC-7H4 solutions and were found to be well represented by the MPL constitutive equation. The friction factor times Reynolds number values were measured in the transition region for which previous measurements were unavailable. Good agreement was found between experiment and calculation thus confirming the validity of the analysis.