<< Click to display Table of content >> Navigation:  Introduction > Le moteur de calcul de MithraSIG

The MITHRA simulation engine, developed by the CSTB, is based on a powerful search algorithm of the paths between sources and receivers in a complex urban environment. The paths consist of direct, diffracted and reflected rays. Because of their design, the algorithms used are suitable for forecasting both in an enclosed environment such as a city centre with a high construction density, in an open environment with vast spaces between buildings and also in mountainous areas where the ground contours affect propagation. The engine has been completely revised and optimized in version 3 of MITHRA-SIG.

The simulation of the propagation of acoustic waves uses powerful algorithms based on asymptotic methods such as adaptive beam propagation. A geometric simulation engine computes the entire contributions between the sources (road, rail and industry) and the observation points (a meshing of receivers on the zone concerned). A physical simulation engine computes the transfer functions associated with the geometric contribution, taking into account the emission spectrum of the sources.

The geometric engine takes into account specular reflections from vertical surfaces and from the ground, and diffractions (according to the uniform theory of diffraction) from horizontal edges (e.g. roofs of buildings) and, possibly vertical edges too.

In order to handle the high volumes of simulated data (on the scale of a city, for example), the geometric modeling used is 2.5D type (of the contours and elevations). The geometric engine is mainly configured by the maximum propagation distance and the order of vertical reflection and diffraction. The geometric contributions computed in 2.5D are converted by the geometric engine into a set of 3D contributions, put through the physical simulation engine.

The physical simulation engine computes noise propagation according to the requirements of current regulations, taking into account the effect of meteorological conditions. The simulation methods in the software are NMPB-2008, NMPB-96, Harmonoise, ISO9613 and CNOSSOS-EU.