NEURAL NETWORK PREDICTIONS OF STUD WALL SOUND INSULATION

  • YEAR
    2000
  • AUTHORS
    Fricke, Fergus
    Coomes, Jill
  • CATEGORIES
    2000 Conference Papers
    Technology and construction

Extract

Using the results of acoustic laboratory tests, neural network analysis has been used to predict the sound
transmission class (STC) for different types of drywall constructions. The basic parameters – stud frame type and
size, mass of wall construction, the inclusion, type and width of any cavity absorption, overall partition width,
minimum sheet lining thickness, and the difference in sheet lining thickness from one side to the other as well as
the inclusion of lining vibration isolation – were used as inputs to the first layer of the neural network. The
results obtained were highly encouraging with successful neural network designs achieving predictions for STC
values within a similar range to those determined by a number of acoustic laboratories, for comparable wall
constructions. A clearer indication of the efficacy of cavity insulation, stud type and vibration were also
obtained from the analysis.

PDF

Download
To top