P-arch
Not a member yet
966 research outputs found
Sort by
Sviluppo di un codice per la previsione delle proprieta di fouling e slagging di carboni: parte prima
Get PDFCRS4 TR 02In questo documento viene inizialmente descritto il modello fisico-matematico su cui è basato l’algoritmo per la descrizione delle distribuzioni tridimensionali di grandezza e composizione chimica delle inclusioni minerarie per ogni classe di grandezza delle polveri di carbone. Carbone che, a sua volta, è descritto da un insieme di dati sperimentali che vanno dai dati ASTM (American Society for Testing and Materials) ai molto più dettagliati dati CSSEM (Computer Controlled Scanning Electron Microscopy). Nella seconda parte del documento viene presentato l’algoritmo di calcolo e lo schema generale del codice, scritto in Fortran 90. Successivamente è presentata una descrizione dettagliata del codice numerico, delle strutture dei dati di input e output, e del suo utilizzo. Di ogni Subroutine viene data un’esplicita descrizione del suo ruolo con frequenti richiami alla parte modellistica e del suo flusso di dati in input/output. Nella sezione dei risultati sono presentati alcuni degli output che è possibile ottenere dell’esecuzione del codice CARBCO. Sono riportati esempi di risultati ottenuti a partire da input sperimentali caratterizzanti il carbone Ashland. Un’interpretazione critica dei risultati è anch’essa presentata. Nella sezione finale sono riportate le conclusioni e un breve sommario
Configuring and installing debian gnu/linux on a 64 bit hp itanium2 architecture
Get PDFThis document describes the work performed to install the Debian GNU/Linux operating system on a Hewlett-Packard zx6000 workstation, based on the IntelItanium2 (McKinley) processor
Interactions model and code generation for J2ME applications
Get PDFThe Java2 Micro Edition (J2ME) platform can really be considered an emerging standard for new generation embedded software. The extent of business opportunities in the domain of small-device software is foreseen to become huge in few years. Though, developing new applications in J2ME is not as easy as developing for desktop computers or for the web. This article introduces a practical methodology
aimed to automatically generate a software prototype starting from an abstract description which defines the dialogue between the user and the application by means of a device independent and abstract description. We will show how an agenda application for cellular phones can be described by means of a visual language called PLANES and present how the personal agenda prototype is implemented by an
appropriate generation tool
Hierarchical higher order face cluster radiosity
Get PDFWe present an algorithm for simulating diffuse interreflection in scenes composed of highly tessellated objects. The method is a higher order extension of the face cluster radiosity technique. It combines face clustering, multiresolution visibility, vector radiosity, and higher order bases with a modified progressive shooting iteration to rapidly produce visually continuous solutions with limited memory requirements. The output of the method is a vector irradiance map that partitions input models into areas where global illumination is well approximated using the selected basis. The OpenGL register combiners extension can be used to render illuminated models directly from the vector irradiance map, exploiting hardware acceleration for computing vertex radiosity on commodity graphics boards
Benchmark calculation of the MEGAPIE target (M1)
Get PDFThe benchmark calculations performed by CRS4 with Star-CD on a reference geometry of the
MEGAPIE target are presented in this report (benchmark M1). Scope of the benchmark is a
comparison of the results obtained by the various partners involved in the MEGAPIE project using
different codes and turbulence modelling approaches.
The considered target geometry is the one with the final part of the guide tube slanted at an angle of
about 9 degrees. The Pb-Bi flow in the last 2150 mm of the target have been simulated, including
the calculation of the thermal field in all the solid structures (window, hull and flow guide). Due to
geometrical symmetry, only half of the real domain was considered. Turbulence was simulated
using a Chen k-ε model, combined with a Two-layer model in the most critical near-wall regions
(window and flow guide in the spallation region) and with Wall Functions along the riser and the
down-comer. Modified wall functions for low Prandtl number fluids were implemented.
Results are presented for both cases with the beam footprint major axis parallel (benchmark M1.0)
and normal (benchmark M1.1) to the guide-tube slant. In order to estimate the effect of the
variation of the turbulent Prandtl number on the heat exchange, two calculation have been
performed, one with Prt = 0.9 and one using a relationship Prt = f(Ret, Pr), yielding a locally
variable turbulent Prandtl number.
Results show a very complex flow pattern in the spallation region, with 3D vortex structures being
generated in the reversing region and dragged along the rising duct.
In case M1.0 with Prt = 0.9, results show maximum window temperatures of 521 °C and 487 °C in
the external and internal side respectively, with a maximum Pb-Bi temperature of 486 °C located
nearby the window centre. The maximum flow velocity is 1.35 m/s. A significant heat exchange
takes place across the 1.5 mm thick flow guide, causing a mean temperature increase along the
down-comer of about 34 °C. Due to the high Reynolds number of the flow, the effect of using a
variable Prt is limited to near wall regions, where the heat exchange is slightly reduced. The
combination of a lower heat exchange across the flow guide (resulting in a lower temperature
increase of the Pb-Bi along the down-comer) and a worse window cooling yielded a maximum
window temperature of 524°C, namely 3 °C more than in the case with Prt = 0.9.
In case M1.1, maximum window temperatures of 447 °C and 414 oC were found using Prt = 0.9
with a maximum Pb-Bi temperature of 423 °C located in the central part of the spallation region.
Using a variable Prt, window temperatures increased of about 2 °C while a 1 °C lower maximum
Pb-Bi temperature was found
A 1D numerical study of the salt diffusion in a solar pond using finite difference tecniques
No full textA simple one dimensional (1D) model for the study of the salt diffusion processes in a solar pond is presented together with the implementation in a computer program using standard finite differences technique. The convective motions are not explicitly taken into account in this model and comments on their importance are presented. Albeit its simplicity the approach here reported can shed light on the basic effects taking place in the dynamics of a solar pond. The importance of the temperature profile and of the Soret effect is discussed
Development of a non-adiabatic premixed model into the Ares combustion code
No full textThe present report somehow represents the summa of the activities which have been carried out during the third and conclusive year of the project. In the necessity of modifying the original goal of the present task, has emerged. In fact, the research work carried out in has shown that the original adiabatic formulation of the TFC model was not sufficient to achieve a qualitatively and quantitatively fair description of the fluid dynamic fields (velocity and temperature) inside a combustion chamber whenever the energy diffusion phenomena cannot be neglected, which is the case in the actual industrial burners. In the same report it was discussed the how the TFC model formulation could be modified in order to take into account the diffusion of thermal energy, and the non adiabatic version of the TFC model available in the code Fluent was tested against the ENSMA experimental combustor test case. The obtained results were such to make advisable to reformulate the original conclusive task of the project in order to implement the non adiabatic version of the TFC model into the codes Ares
Multi-frequency analysis of electrical ground conductivity data
Get PDFThis document presents the general problem of electromagnetic inversion in the frequency
domain. The 1-dimensional analysis of a stratified medium is performed at different frequencies;
the data measured on the surface (the admittance or the apparent electric conductivity) are compared
with the solution of the system obtained from a Lagrangian formulation which minimizes
a suitable cost function. The information acquired at different frequencies is integrated in the inversion
procedure; two approaches are presented and compared: the first method independently
solves the problem for each frequency and then integrates the contributions, while the second
method solves a global problem which involves all the contributions at the same time. The conjugate
gradient algorithm has been used as iterative minimization strategy for both procedures.
Numerical examples validate this analysis
