Predicting HCl concentrations in fire enclosures using an HCl decay model coupled to a CFD-based fire field model

The amount of atmospheric hydrogen chloride (HCl) within fire enclosures produced from the combustion of chloride-based materials tends to decay as the fire effluent is transported through the enclosure due to mixing with fresh air and absorption by solids. This paper describes an HCl decay model, typically used in zone models, which has been modified and applied to a computational fluid dynamics (CFD)-based fire field model. While the modified model still makes use of some empirical formulations to represent the deposition mechanisms, these have been reduced from the original three to two through the use of the CFD framework. Furthermore, the effect of HCl flow to the wall surfaces on the time to reach equilibrium between HCl in the boundary layer and on wall surfaces is addressed by the modified model. Simulation results using the modified HCl decay model are compared with data from three experiments. The model is found to be able to reproduce the experimental trends and the predicted HCl levels are in good agreement with measured values

Dynamic model for metal cleanness evaluation by melting in a cold crucible

Melting of metallic samples in a cold crucible causes inclusions to concentrate on the surface owing to the action of the electromagnetic force in the skin layer. This process is dynamic, involving the melting stage, then quasi-stationary particle separation, and finally the solidification in the cold crucible. The proposed modeling technique is based on the pseudospectral solution method for coupled turbulent fluid flow, thermal and electromagnetic fields within the time varying fluid volume contained by the free surface, and partially the solid crucible wall. The model uses two methods for particle tracking: (1) a direct Lagrangian particle path computation and (2) a drifting concentration model. Lagrangian tracking is implemented for arbitrary unsteady flow. A specific numerical time integration scheme is implemented using implicit advancement that permits relatively large time-steps in the Lagrangian model. The drifting concentration model is based on a local equilibrium drift velocity assumption. Both methods are compared and demonstrated to give qualitatively similar results for stationary flow situations. The particular results presented are obtained for iron alloys. Small size particles of the order of 1 μm are shown to be less prone to separation by electromagnetic field action. In contrast, larger particles, 10 to 100 μm, are easily “trapped” by the electromagnetic field and stay on the sample surface at predetermined locations depending on their size and properties. The model allows optimization for melting power, geometry, and solidification rate.

Carbon fluxes and pelagic ecosystem dynamics near two western Antarctic Peninsula Adélie penguin colonies: an inverse model approach

An inverse food-web model for the western Antarctic Peninsula (WAP) pelagic food web was constrained with data from Palmer Long Term Ecological Research (PAL-LTER) project annual austral summer sampling cruises. Model solutions were generated for 2 regions with Adelie penguin Pygoscelis adeliae colonies presenting different population trends (a northern and a southern colony) for a 12 yr period (1995-2006). Counter to the standard paradigm, comparisons of carbon flow through bacteria, microzooplankton, and krill showed that the diatom-krill-top predator food chain is not the dominant pathway for organic carbon exchanges. The food web is more complex, including significant contributions by microzooplankton and the microbial loop. Using both inverse model results and network indices, it appears that in the northern WAP the food web is dominated by the microbial food web, with a temporal trend toward its increasing importance. The dominant pathway for the southern WAP food web varies from year to year, with no detectable temporal trend toward dominance of microzooplankton versus krill. In addition, sensitivity analyses indicated that the northern colony of Adelie penguins, whose population size has been declining over the past 35 yr, appears to have sufficient krill during summer to sustain its basic metabolic needs and rear chicks, suggesting the importance of other processes in regulating the Adelie population decline.

The Development of a Two-Dimensional Transient Catalyst Model for Direct Injection Two-Stroke Applications.

This paper describes the development of a two-dimensional transient catalyst model. Although designed primarily for two-stroke direct injection engines, the model is also applicable to four-stroke lean burn and diesel applications. The first section describes the geometries, properties and chemical processes simulated by the model and discusses the limitations and assumptions applied. A review of the modeling techniques adopted by other researchers is also included. The mathematical relationships which are used to represent the system are then described, together with the finite volume method used in the computer program. The need for a two-dimensional approach is explained and the methods used to model effects such as flow and temperature distribution are presented. The problems associated with developing surface reaction rates are discussed in detail and compared with published research. Validation and calibration of the model is achieved by comparing predictions with measurements from a flow reactor. While an extensive validation process, involving detailed measurements of gas composition and thermal gradients, has been completed, the analysis is too detailed for publication here and is the subject of a separate technical paper.

The Validation of a Two-Dimensional Transient Catalyst Model for Direct Injection Two-Stroke Applications

This paper describes the detailed validation of a computer model designed to simulate the transient light-off in a two-stroke oxidation catalyst. A plug flow reactor is employed to provide measurements of temperature and gas concentration at various radial and axial locations inside the catalyst. These measurements are recorded at discrete intervals during a transient light-off in which the inlet temperature is increased from ambient to 300oC at rates of up to 6oC/sec. The catalyst formulation used in the flow reactor, and its associated test procedures, are then simulated by the computer and a comparison made between experimental readings and model predictions. The design of the computer model to which this validation exercise relates is described in detail in a separate technical paper. The first section of the paper investigates the warm-up characteristics of the substrate and examines the validity of the heat transfer predictions between the wall and the gas in the absence of chemical reactions. The predictions from a typical single-component CO transient light-off test are discussed in the second section and are compared with experimental data. In particular the effect of the temperature ramp on the light-off curve and reaction zone development is examined. An analysis of the C3H6 conversion is given in the third section while the final section examines the accuracy of the light-off curves which are produced when both CO and C3H6 are present in the feed gas. The analysis shows that the heat and mass transfer calculations provided reliable predictions of the warm-up behaviour and post light-off gas concentration profiles. The self-inhibition and cross-inhibition terms in the global rate expressions were also found to be reasonably reliable although the surface reaction rates required calibration with experimental data.

Gas-liquid two phase flow through a vertical 90deg elbow bend

Pressure drop data are reported for two phase air-water flow through a vertical to horizontal 90° elbow bend set in 0.026 m i.d. pipe. The pressure drop in the vertical inlet tangent showed some significant differences to that found for straight vertical pipe. This was caused by the elbow bend partially choking the inflow resulting in a build-up of pressure and liquid in the vertical inlet riser and differences in the structure of the flow regimes when compared to the straight vertical pipe. The horizontal outlet tangent by contrast gave data in general agreement with literature even to exhibiting a drag reduction region at low liquid rates and gas velocities between 1 and 2 m s -1. The elbow bend pressure drop was best correlated in terms of le/d determined using the actual pressure loss in the inlet vertical riser. The data showed a general increase with fluid rates that tapered off at high fluid rates and exhibited a negative pressure region at low rates. The latter was attributed to the flow being smoothly accommodated by the bend when it passed from slug flow in the riser to smooth stratified flow in the outlet tangent. A general correlation was presented for the elbow bend pressure drop in terms of total Reynolds numbers. A modified Lockhart-Martinelli model gave prediction of the data.

Comparison of two construction algorithms for local model networks

The divide-and-conquer approach of local model (LM) networks is a common engineering approach to the identification of a complex nonlinear dynamical system. The global representation is obtained from the weighted sum of locally valid, simpler sub-models defined over small regions of the operating space. Constructing such networks requires the determination of appropriate partitioning and the parameters of the LMs. This paper focuses on the structural aspect of LM networks. It compares the computational requirements and performances of the Johansen and Foss (J&F) and LOLIMOT tree-construction algorithms. Several useful and important modifications to each algorithm are proposed. The modelling performances are evaluated using real data from a pilot plant of a pH neutralization process. Results show that while J&F achieves a more accurate nonlinear representation of the pH process, LOLIMOT requires significantly less computational effort.

Fast automatic two-stage nonlinear model identification based on the extreme learning machine

It is convenient and effective to solve nonlinear problems with a model that has a linear-in-the-parameters (LITP) structure. However, the nonlinear parameters (e.g. the width of Gaussian function) of each model term needs to be pre-determined either from expert experience or through exhaustive search. An alternative approach is to optimize them by a gradient-based technique (e.g. Newton’s method). Unfortunately, all of these methods still need a lot of computations. Recently, the extreme learning machine (ELM) has shown its advantages in terms of fast learning from data, but the sparsity of the constructed model cannot be guaranteed. This paper proposes a novel algorithm for automatic construction of a nonlinear system model based on the extreme learning machine. This is achieved by effectively integrating the ELM and leave-one-out (LOO) cross validation with our two-stage stepwise construction procedure [1]. The main objective is to improve the compactness and generalization capability of the model constructed by the ELM method. Numerical analysis shows that the proposed algorithm only involves about half of the computation of orthogonal least squares (OLS) based method. Simulation examples are included to confirm the efficacy and superiority of the proposed technique.