|Program name||Package id||Status||Status date|
|Package ID||Orig. computer||Test computer|
|NEA-1886/02||MAC,Linux-based PC,PC Windows||PC Windows|
The Fortran subroutine package PENGEOM and the associated graphical user interface PenGeomJar constitute a complete set of tools for handling complex quadric geometries in Monte Carlo simulations of radiation transport. The material structure where radiation propagates is assumed to consist of homogeneous bodies limited by quadric surfaces. The PENGEOM subroutines (a subset of the PENELOPE code) automatically track particles through the material structure, independently of the details of the physics models adopted to describe the interactions. These subroutines are designed for detailed simulation schemes, where all individual interactions of the transported particles are simulated sequentially. They also work with mixed (class II) schemes for high-energy charged particles, where the effect of soft interactions is described by the random-hinge method. The tracking algorithm and the definition of the geometry are tailored to optimize simulation speed. The Java graphical user interface allows editing and debugging the geometry definition file, as well as visualizing the material structure.
Some bugs related to the graphic capabilities have been corrected in version NEA-1886/02.
Particle trajectories are generated following a detailed or class II simulation scheme. Intersections of rays with limiting quadric surfaces are calculated analytically. Because of round-off errors, the effective resolution worsens when the distance to the origin of coordinates increases. The impact of these errors is reduced by considering that limiting surfaces are fuzzy. The PENGEOM subroutines are capable of resolving a sphere of unit radius located at a distance of 107 length units from the origin.
The connection between the steering main program and the tracking routines is through a Fortran module, which contains the state variables of the transported particle. The generation of a particle trajectory reduces to a call to subroutine LOCATE, which initializes the track, and a sequence of calls to subroutine STEP, which moves the particle across the material structure.
The running time much depends on the complexity of the geometrical structure. The most complex example provided is an anthropomorphic phantom with 264 surfaces and 169 bodies. The rendering of 2D images of that phantom is almost instantaneous, while the generation of 3D images with 1680 × 1050 pixels takes about 25 seconds on an Intel Core I7-3520M CPU with Windows 7.
|Package ID||Status date||Status|
|NEA-1886/02||08-JAN-2016||Tested at NEADB|
|Package ID||Computer language|
Servicio de Radiofísica y P.R., Hosp. Univ. Virgen de las Nieves,
Avda. de las Fuerzas Armadas 2, 18014 Granada, Spain
Institut für Theoretische Physik, Goethe-Universität Frankfurt,
Max-von-Laue-Straße 1, 60438 Frankfurt am Main, Germany
Departamento de Ciencias Físicas, Universidad de La Frontera,
Av. Francisco Salazar, 01145 Temuco, Chile
Artur Carnicer and Francesc Salvat
Facultat de Física, Universitat de Barcelona,
Diagonal 647, 08028 Barcelona, Spain
Antonio M. Lallena
Departamento de Física Atómica, Molecular y Nuclear,
Universidad de Granada, 18071 Granada, Spain
Keywords: Monte Carlo, constructive quadric geometry, geometry visualisation, particle transport, ray-tracing.