Determining Noise in an Aggregates Plant Using Functional Statistics

ABSTRACT

A key issue to be taken into account in any mining operation or premises is workers’ health and safety. Noise has negative repercussions on workers’ health, which is why regular inspections of noise levels are necessary to ensure that noise is less than legally permitted limits. In our study we statistically analyzed functional data with a view to determining noise levels in an aggregates plant. Data were captured by a sound level meter that indicated equivalent sound pressure levels and peaks and the octave bands for a specific range of frequencies. Unlike the methods habitually used to construct noise maps (which only permit one of these parameters to be determined for each point), the methods applied in our research enable the noise level to be calculated at each point for the entire spectrum of frequencies picked up by a sound level meter. For each point we obtain, rather than a single value, a noise curve reflecting the entire range of frequencies detected by the sound level meter. This enables us to better understand the effects of noise and to determine its origin.

Key Words:

• functional data
• noise
• sound level meter
• kriging
• support vector machines

ACKNOWLEDGMENT

J. M. Matías's research is funded by the Spanish Ministry of Education and Science under Grant. No. MTM2008-03010.