1,720,969 research outputs found
Computational optimization of a subsonic compressible gas Venturi's ejector
No full textThe paper deals with a 3-D numerical simulation and validation against industrial measurements of turbulent frozen reacting flow in a subsonic compressible gas Venturi’s ejector used as “fluid dynamic engine” for external flue gas recirculation in a state-of-the-art “annular shaft” lime kiln. Higher stagnation thermodynamics parameters of the ejector hot gas primary stream permit the avoidance of the condensing temperature window of compounds such as K2O and Na2O, and KCl and NaCl that produce sticky builds-up on the ejector’s internal wall. An improved gas dynamics effectiveness allows the maximization of the amount of secondary flue gas stream using much less primary stream mass flow rate. The commercial FluentTM UNS/5 software was used to predict all flow behaviour characteristics inside the original and new Venturi’s ejectors. A reasonable agreement has been found between the computed and experimental flow rate figures of the secondary flue gas stream of the actual functional kiln
Modelling of internal nozzle flow in high pressure water mist injector for fire suppression applications
No full textThe internal flow in a water mist injector for fire suppression applications is investigated, implementing 3D Large Eddy Simulations based on the Volume-of-Fluid methodology. The flow is assumed to be incompressible under isothermal non-reacting conditions. The effect of internal nozzle geometry on the injector behaviour is investigated, for different operating conditions, by modifying the inclined swirling channels. The injector behaviour is mapped varying the injection pressure for each nozzle geometry. Detailed in-nozzle flow analysis is performed
Numerical analysis of regime stability of a water mist pressure swirl atomizer
No full textThe two-phase internal flow in water mist injectors is predicted by 3D numerical simulations, where the gas/liquid interface is captured by the volume of fluid methodology and the effect of turbulence is accounted for by the large eddy simulation approach, assuming incompressible flow under isothermal conditions. The numerical model is validated against experimental data for two large-scale pressure swirl atomizers from the literature, showing a satisfactory agreement. The flow development in real-size atomizers is studied under a wide range of swirling conditions, defined by different injection pressures and swirling channel inclinations. Three internal flow regimes are identified, depending on the behavior of the air core at the discharge hole: an “open” regime, where a stable air core is formed inside the nozzle inducing the so-called hollow-cone liquid jet, a “closed” regime, where the discharge hole is fully occupied by the liquid, and an “unstable” regime characterized by a random switching between the open and closed regimes. The mass flow rate, the pressure field, the liquid phase fraction, and the flux of angular momentum are monitored along the nozzle, and a relation between the swirl number, defined as a nondimensionalization of the flux of angular momentum, at nozzle exit and the air core diameter is found. The analysis of the conditions causing the transition between the open and closed regimes and vice versa suggests that there exists a threshold value of the swirling number at the nozzle exit that defines the working regimes independently on the geometric and operating conditions
The effects of shape and liquid properties on pressure swirl atomizer in-nozzle flow
No full textThe internal flow in pressure swirl atomizers is numerically predicted by performing large eddy simulations and using a volume of fluid approach. The output of the numerical model is validated by comparing it with three databases of experimental measurements obtained on large-scale pressure swirl atomizers available in the open literature. A simplified analytical model previously developed by the authors, which relates the swirl intensity to the thickness of the fluid exiting the nozzle, is used to analyze the flow behavior in three pressure swirl atomizers, with large differences in the injector geometry, the operating conditions, and the fluid thermophysical properties. This simple relationship is found to hold for the three pressure swirl atomizers, with small changes of the parameter that accounts for energy losses, while data obtained with relatively small variations of the injector geometry are found to collapse on the same curve. The effects of operating conditions and fluid thermophysical properties on this relation are found to be irrelevant
Numerical investigation of natural convection effects on sessile and pendant evaporating drops
No full textThis study investigates the effect of natural convection on the evaporation of sessile and pendant drops, accounting for buoyancy forces arising from variations in gas density due to non-uniform distributions of vapor concentration and temperature in the surrounding gas mixture. The governing conservation equations are solved using the simulation software Ansys Fluent on two-dimensional axisymmetric grids under steady-state conditions and assuming that the drops are spherical caps. A range of drop sizes is considered to span Grashof numbers relevant to typical applications. Three evaporating fluids (n-octane, ethanol, and water) with molar masses higher or lower than that of air are analyzed, and helium as the inert species was used to illustrate the effect of buoyancy in evaporation. The impact of surface wettability is examined by considering a range of contact angles between 30° and 150°. The influence of natural convection on sessile and pendant drops is quantified, highlighting the role of gas mixture composition. A novel correlation is derived, with the scope to correct the evaporation rate predicted by the diffusion-driven analytical model, and to account for the buoyancy force in the case of both sessile and pendant drops vaporizing on substrates with different wettability
Modeling hemodynamics in intracranial aneurysms: Comparing accuracy of CFD solvers based on finite element and finite volume schemes
Get PDFImage-based computational fluid dynamics (CFD) has shown potential to aid in the clinical management of intracranial aneurysms, but its adoption in the clinical practice has been missing, partially because of lack of accuracy assessment and sensitivity analysis. To numerically solve the flow-governing equations, CFD solvers generally rely on 2 spatial discretization schemes: finite volume (FV) and finite element (FE). Since increasingly accurate numerical solutions are obtained by different means, accuracies and computational costs of FV and FE formulations cannot be compared directly. To this end, in this study, we benchmark 2 representative CFD solvers in simulating flow in a patient-specific intracranial aneurysm model: (1) ANSYS Fluent, a commercial FV-based solver, and (2) VMTKLab multidGetto, a discontinuous Galerkin (dG) FE-based solver. The FV solver's accuracy is improved by increasing the spatial mesh resolution (134k, 1.1m, 8.6m, and 68.5m tetrahedral element meshes). The dGFE solver accuracy is increased by increasing the degree of polynomials (first, second, third, and fourth degree) on the base 134k tetrahedral element mesh. Solutions from best FV and dGFE approximations are used as baseline for error quantification. On average, velocity errors for second-best approximations are approximately 1 cm/s for a [0,125] cm/s velocity magnitude field. Results show that high-order dGFE provides better accuracy per degree of freedom but worse accuracy per Jacobian nonzero entry as compared with FV. Cross-comparison of velocity errors demonstrates asymptotic convergence of both solvers to the same numerical solution. Nevertheless, the discrepancy between underresolved velocity fields suggests that mesh independence is reached following different paths
CFD simulations of a two-phase ejector for transcritical CO2 cycles applied to supermarket refrigeration systems
No full textThe use of carbon dioxide (CO2), as a working fluid for large refrigeration systems, has grown tremendously in recent years. Factors such as its low cost, easy accessibility and environmentally friendly characteristics compared to HFCs and HCFCs, has made CO2a viable alternative. To efficiently operate with CO2, the thermodynamic cycle needs high-pressure levels that can easily exceed the critical point due to the low critical temperature. By replacing conventional expansion valves with ejectors, the thermodynamic losses of the high-pressure throttling are mitigated, and the overall system performance is improved. To design and efficiently control the whole cycle, a thorough comprehension of the ejector fluid dynamics is mandatory. In this work, Computational Fluid Dynamics (CFD) is used to thoroughly investigate such a device. The employed CFD solver uses a modified form of the Homogenous Relaxation Model (HRM) to deal with two-phase flows in a non-thermodynamic equilibrium state, per Colarossi et al. (2012). Preliminary numerical results for an ejector in supermarket refrigeration system operating conditions are presented and discussed
Assessment of a discontinuous Galerkin method for the simulation of the turbulent flow around the DrivAer car model
Get PDFThe turbulent flow over the DrivAer fastback model is here investigated with an orderadaptive discontinuous Galerkin (DG) method. The growing need of high-fidelity flow simulations for the accurate determination of problems, e.g., vehicle aerodynamics, promoted research on models and methods to improve the computational efficiency and to bring the practice of Scale Resolving Simulations (SRS), like the large-eddy simulation (LES), to an industrial level. An appealing choice for SRS is the Implicit LES (ILES) via a high-order DG method, where the favourable numerical dissipation of the space discretization scheme plays directly the role of a subgrid-scale model. Implicit time integration and the p-adaptive algorithm reduce the computational cost allowing a high-fidelity description of the physical phenomenon with very coarse mesh and moderate number of degrees of freedom. Two different models have been considered: (i) a simplified DrivAer fastback model, without the rear-view mirrors and the wheels, and a smooth underbody; (ii) the DrivAer fastback model, without rear-view mirrors and a smooth underbody. The predicted results have been compared with experimental data and CFD reference results, showing a good agreement
Numerical simulations of internal flow in an aircraft engine pressure swirl atomizer
No full texthe internal flow and the liquid structure at the nozzle exit are investigated in a typical pressure swirl atomizer for aeroengine applications under an isothermal nonreacting environment using a two-phase flow modeling according to the volume-of-fluid numerical methodology. A parametrical analysis is performed to investigate the effect of operating conditions on the injector performances and the characteristics of the liquid jet exiting the nozzle. The liquid film thickness at the nozzle exit and the injector discharge coefficient predicted by the simulations are compared against commonly used correlations available in the literature. The influence of the turbulence model on the injector performances is also analyzed using three different models: the renormalization group k-e model, the Reynolds stress model, and the large-eddy simulation model
Sessile and suspended droplets on a biphilic surface in the presence of natural convection: Experimental studies and modeling
No full textNew experimental and modeling results, referring to heating and evaporation of sessile and pendant water droplet on a biphilic surface, are presented. Two modeling approaches are used: one based on the previously developed variable density model in which the droplet shape in the presence of gravity is described by the Bashforth-Adams equation, and the other based on ANSYS Fluent. It is shown that the results predicted by both approaches almost coincide in the absence of gravity, which can be considered as verification of both approaches. The predictions of both approaches are shown to be close to experimental results for pendant droplets. For sessile droplets, however, both approaches tend to under-predict experimental data. The difference in model predictions, taking and not taking into account the effect of natural convection, is shown not to exceed 3% for the experimental conditions under consideration
- …
