P-arch
Not a member yet
966 research outputs found
Sort by
Matching techniques to compute image motion
This paper describes a thorough analysis of the pattern matching techniques used to compute image motion from a sequence of two or more images. Several correlation/distance measures are tested, and problems in displacement estimation are investigated. As a byproduct of this analysis, several novel techniques are presented which improve the accuracy of flow vector estimation and reduce the computational cost by using filters, multi-scale approach and mask sub-sampling. Further, new algorithms to obtain a sub-pixel accuracy of the flow are proposed. A large amount of experimental tests have been performed to compare all the techniques proposed, in order to understand which are the most useful for practical applications, and the results obtained are very accurate, showing that correlation-based flow computation is suitable for practical and real-time applications.247–260Pubblicat
A thermal fluid-dynamic transient analysis of the EADF down-comer channel
In this work a numerical simulation of the downcomer channel of the Energy Amplifier Demonstration Facility (EADF) [1, 2] is presented. The simulation is fully three-dimensional (3D) and is focused on a transient analysis. All relevant heat transfer phenomena are taken into account. Starting from the nominal power configuration, we have simulated the response of the system to a power shutdown of the core for a period of 60 s. The core shutdown is simulated imposing a linear variation of the inlet flow temperature from 400°C to 305°C in 10 s. The simulation shows the evolution of the thermal stratification outside the IHX and the evolution of the IHX operation
Prima esperienza di sviluppo di un sistema informativo geografico della piana di Capoterra (Sardegna)
Le informazioni acquisite in occasione di studi idrogeologici precedentemente svolti sulla piana alluvionale di Capoterra sono stati organizzati in un sistema informativo geografico, la cui prima esperienza ha riguardato lo studio della vulnerabilità intrinseca dell’acquifero superficiale all’inquinamento.The information acquire in occasion of hydrogeological studies previously carried out in the alluvial plain of Capoterra has stayed organized in a geographical information system, whose first experience has concerned the study of the intrinsic vulnerability of the phreatic aquifer to the pollution
Analysis of seawater intrusion phenomena in the Korba coastal aquifer of Tunisia
A geographic information system and a three-dimensional coupled variable density and saturation numerical model are implemented for the Korba coastal aquifer of northeastern Tunisia, and preliminary simulations are performed to investigate seawater intrusion phenomena. The GIS provides an essential platform for data management, enabling the organization and merging of a large volume of data that has been collected in often ad hoc manner from diverse sources over many years. A critical assessment of data quality is provided and the usefulness of GIS and modeling tools is demonstrated, with an aim to encourage more directed and continuous monitoring and characterization of important parameters and processes involved in the contamination phenomena. This approach is currently being extended to two other coastal aquifers, in Sardinia (Italy) and Sahel
(Morocco).51-5
Time domain analysis of ship motion and wave loads by boundary integral equations
In this presentation, we like to discuss some aspects of a more general computational algorithm for the prediction of ship motion and loads induced by the interaction with wave systems. In particular, unlikely the more conventional models in frequency domain, we attack the problem by a time domain formulation. The purpose is twofold. First, within the framework of a linear analysis, the ship response function to a general wave excitation can be numerically determined by a transient test (i.e. the interaction with a wave pulse compact in time). In this way, a substantial saving of computational time with respect to the existing algorithm is achieved. Second, a time domain modeling is intrinsically prone to deal with the fully nonlinear problem or, at least, to recover some nonlinear effects
Mechanistic investigation of electric field-activated self-propagating reactions: experimental and modeling studies
The mechanism of electric field-activated self-propagating reactions is investigated using the
combustion front quenching technique. In particular, previously published experimental results obtained through the Field Assisted Combustion Synthesis (FACS) of b-SiC, TaC, Ti3Al and B4C-TiB2 are re-examined and compared. Pre-combustion and combustion stages involved during synthesis wave propagation are postulated for all systems. Subsequently, modeling results aimed at simulating the process where an electric field-activated self-propagating reaction takes place are presented. In particular, a one-dimensional model of FACS technique is developed to simulate the rapid quenching of the reaction during its progress as the applied field is turned off. A rate expression which accounts for the influence of temperature, particle size, compaction density, reactant stoichiometry, and inert content is included in the model
Development, implementation and numerical tests of implicit methods
The goal of this work was to implement an implicit version of Karalis, a Navier-Stokes equations solver. Besides the Spalart & Allmaras equation is formulated in an implicit way. In this paper the first part is devoted to the implementation details of the implicit Navier-Stokes equations while the latter to the Spalart & Allmaras equation. The routines are then briefly described
Numerical simulations on the TEFLU sodium jet experiment using the CFD code Karalis
The TEFLU experiments were performed at the Karlsruhe Forschungszentrum in order to investigate the behaviour of a low-Prandtl number jet under various flow condition, from forced flow to purely buoyant. Here, numerical simulations are presented and results compared with the experimental data. The computation are made within the Benchmark Working Group activities in order to test the capabilities of CFD codes to simulate Heavy Liquid Metal flows with heat transfer. The simulation are performed with the CFD code Karalis
The CFD code karalis
Karalis is a paralle MPI, Finite-Volume, multiblock CFD code which solves the fully compressible Euler and Navier-Stokes equations where all couplings between dynamics and thermodynamics are allowed. This the most general mathematical model for all fluid flows. The code solves the coupled system of continuity, momentum and full energy equation for the velocity components, pressure and temperature. Once, u, v, w, p are and T are updated, arbitrary thermodynamics is supplied. The second order Roe’s upwind TVD scheme is used to compute convective fluxes through the Finite-Volume cell interfaces. A V-cycle Coarse Grid Correction Multi-Grid algorithm is used, together with a 5-stage Runge-Kutta explicit time-marching method, to accelerate convergence to a steady state. This formulation, typical of aerodynamic flows, shows an eccellent efficiency even for incompressible flows as well as for flows of incompressible fluids (typically buoyancy flows), once equipped with a preconditioner. Merkel’s preconditioner has been chosen because it can be easily formulated for arbitrary equations of state given as a functional relation of two independent thermodynamic variables (typically the pressure p and the temperature T), or even in tabular form, read in as an input file and used with bilinear interpolation. Karalis implement two among the most popular turbulence models, namely the one-equation model by Spalart and Allmaras and the two-equations model by Wilcox, the k-ω model, which allow a good compromise between accuracy, robustness and stability of turbulent calculations. Code validation is presented for some typical benchmark test cases of incompressible fluid dynamics. Comparison with solutions obtained with a few popular commercial CFD codes is also presented