Fouad Ammouri

Soot formation effects of oxygen concentration in the oxidizer stream of laminar coannular nonpremixed methane/air flames

Abstract This experimental investigation analyzes the soot formation effects of oxygen concentration in the oxidizer stream (O 2 + N 2 ) ventilating laminar jet nonpremixed methane flames. The base flame incorporates air as the oxidizer; two additional flames, with respective oxygen concentrations of 50% and 100% in the ventilating coflow, are also examined. The microstructure of soot collected from selected flame locations is determined combining thermophoretic sampling and transmission electron microscopy. A laser-light extinction technique is employed along with tomographic inversion to measure the soot volume fraction distributions within the three flames. The results indicate that soot surface growth and oxidation rates in the methane/50% oxygen flame are higher compared to the respective rates in the methane/air base flame. The rate of soot inception becomes stronger with increasing oxygen content in the oxidizer stream. Soot yields diminish with increasing oxygen concentration, as do luminous flame spatial dimensions. Soot aggregate data on the soot annulus suggest a higher degree of agglomeration under oxygen-enriched conditions. Finally, the fractal dimensions of selected soot aggregate samples are measured to be 1.64 (methane/air flame) and 1.65 (methane/50% oxygen flame), being similar to previously published values for carbonaceous soot.

Stability and asymptotic observers of binary distillation processes described by nonlinear convection/diffusion models

Distillation column monitoring requires shortcut nonlinear dynamic models. On the basis of a classical wave-model and time-scale reduction techniques, we derive a one-dimensional partial differential equation describing the composition dynamics where convection and diffusion terms depend non-linearly on the internal compositions and the inputs. The Cauchy problem is well posed for any positive time and we prove that it admits, for any relevant constant inputs, a unique stationary solution. We exhibit a Lyapunov function to prove the local exponential stability around the stationary solution. For a boundary measure, we propose a family of asymptotic observers and prove their local exponential convergence. Numerical simulations indicate that these convergence properties seem to be more than local.

Input/output transfer models of binary distillation columns derived from convection-diffusion partial differential equations

Purity control in distillation columns often relies on linear models whose parameters (gains, delays, time constants) may be hard to fit satisfyingly. In this paper, we linearize a convection-diffusion model inspired from wave-models, in the case of vanishing diffusion. We obtain a linear transfer model which is explicitly parameterized with the usual columns operating parameters. This model consists in gains and pure delays only. It is derived from an unusual asymptotic expansion (Laurent series with exponentially precise terms) versus the diffusion coefficient. We validate such linear transfer models in simulation and also on real plants data for cryogenic air separation units. Similar transfer models can be constructed for other cases of binary distillation.

On-line Calculation of Critical Characteristics for Cryogenic Storage Vessels

Cryogenic products, such as Oxygen, Nitrogen or Argon, are commonly delivered to customers by the supplier. To do so, the supplier is responsible for managing the inventory of product at the customer site, in most cases. It is important to have access to an accurate estimation of the quantity of product stored at any time in cryogenic vessels to avoid run-outs for the customer and to optimize the product deliveries, which means saving money and reducing pollution due to trucks fuel consumption. Former studies showed that an accurate mass estimation can be made from differential pressure measures between the bottom and the top of the vessel, the pressure on top, and required vessel characteristics. Differential pressure measurements are installed on cryogenic tanks and the data is available remotely through a remote telemetry unit (RTU). Often, the tank characteristics are unknown or only roughly estimated limiting the accuracy of the mass estimation in the tank. This paper presents an on-line method to estimate the critical required tank characteristics: radius, total height (for vertical tanks), total length (for horizontal tanks), bottom height, and maximum liquid height. This method relies on a physical model describing the vessel during a filling and providing a relationship between the delivered mass of product, the pressure measures and the vessel characteristics.

Coupling a Discrete Transfer Method to a View Factors Technique for Radiative Heat Transfer Modeling in Industrial Furnaces

Air Liquide has been involved in the design of industrial furnaces (glass melting, reheating, aluminum, [[ellipsis]]) for several years. Thanks to that experience, known-how and expertise in modeling such applications have been developed. Dedicated simulation tools — 0D for global heat and mass balance, 1D for the prediction of longitudinal temperature profiles and 3D for detailed analysis — have been built. Each of them is very helpful when used relevantly and offers numerous opportunities at each step of the design of a furnace. In such kind of applications, the temperature levels are very high (up to 2500 K). As a consequence it is very crucial to simulate the radiative heat transfer as accurately as possible. This requires the use of a radiation model that can take into account complex geometries, non-isothermal media and various gas mixture compositions. Very often, three-dimensional simulations are necessary and reduction to smaller dimension problems is difficult or inadequate. The present paper introduces a new radiation model for computing two-dimensionally radiative heat transfer in an industrial furnace with a piecewise distributed load. To reduce the three-dimensional problem to two dimensions, the method consists in coupling the 2D radiation transport equation to a boundary condition based on view factors through an imaginary plane to homogenize the radiative behavior of the load surface. A solution procedure using the discrete transfer method associated to a weighted-sum-of-gray-gases database to deal with absorption and emission of a CO2 -H2 O mixture is proposed. Simulation results are finally compared to an analytical formula and then to a full-3D approach taking into account participating media, non-isothermal and gray walls. All tests show that this model can be used to simulate industrial configurations with a good accuracy.© 2005 ASME