For a long time there is a need in industry of acoustical modeling of rooms. Modeling is necessary for new production room design, the exchange of existing machinery, modernization or expansion of production rooms, changes in production profile or acoustical room adaptation for acoustical work conditions improvement. in such cases modeling quality is essential and thanks to uncertainty analysis it is possible to quantitatively estimate the effect that input parameters value variation has on model behavior. The article presents general rules for sound pressure level prediction uncertainty calculation in a room. By partial uncertainty calculation analysis of input parameters influence on uncertainty prediction an effort was taken to find parameters with biggest influence on the prediction process. As an example an industrial production room is presented which was modeled to predict noise level on a work stands after it was expanded.
Results of computer simulations and measurements in real interior for varying location of sound source and microphone are shown in the paper. A small room with a volume of 47 m^3 was used for this purpose. The objective of measurements and calculations was to determine the sound pressure level and other parameters derived from the room impulse response (T_{30}, EDT, C_{80}, STI) followed by the sensitivity analysis of those parameters to changing the location and orientation of the sound source and the receiver. In order to determine these parameters, the room impulse response was measured using MLS method. Experimental studies have been used to verify the acoustic room model built with use of enhanced radial method algorithms and its sensitivity. That allowed complementary and extended simulation studies on the room acoustic characteristics and finally determination of sensitivity of output parameters to changes of location and orientation of the measurement channel elements.
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.