Project Details
Description
This investigation attempts to reveal general relationships between room geometry and the acoustic properties in concert halls through statistical analysis of the data available in our existing data base. The examined objective acoustic parameters for audience are EDT, ts, LEF, C, G, G(r) (Rate of Spatial Decay [dB/10 m]), and for musicians, EDTP (EDT measured on the podium) and ST1. The acoustic parameters are averaged over frequency and position before the analyses. This year correlation analysis, principal component analysis and multible linear regression analysis have been carried out in order to see if these methods were able to reveal further information on the interaction between room geometry and acoustic properties. Besides, new geometrical parameters were included in the analyses; some were taken from newer literature and some were developed within the project. Also, two general and flexible room models were made in the simulation program ODEON, one rectangular model and one fanshaped model, in which single geometrical parameters were varied in an attempt to verify the regression models. For the acoustic parameters the correlation analysis showed that often many of the geometrical parameters had a good correlation with the acoustic parameters or with the difference between the parameter and its expected value according to diffuse field theory. Furthermore, the principal component analysis revealed complicated interactions among the geometrical parameters. From the simulations in ODEON it was generally found that the altered geometrical parameter showed the same effect on a current acoustic parameters as found from the regression models. The similarity between the ODEON calculations and the regression models was more distinct for acoustic parameters not related to reverberation time (ie parameters for which no value can be predicted via diffuse field theory).
Status | Active |
---|---|
Effective start/end date | 01/01/1992 → … |
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.