UNCERTAINTY ANALYSIS FOR CORONA AUDIBLE NOISE LONG-TERM RATE LEVEL ESTIMATION IN CONTINUOUS MONITORING SYSTEMS IN VICINITY OF OVERHEAD AC POWER LINES

摘要

Corona audible noise (AN) characterizes itself some specific features such as large time variation of its level and spectral structure dependent on the weather conditions as well as on the technical state of the line and its surface contamination, both factors being of random nature. The measurement of corona AN level is often difficult because of its low values, sometimes comparable with the background noise. A quality of the long-term noise indexes estimation might be expressed by uncertainty analysis for the whole measurement system including randomness of weather condition and its affect the phenomenon of corona intensity and sound travelling conditions as well. Using uncertainty propagation law an analysis has been made for an influence of the partial uncertainties. In the uncertainty budget some partial uncertainties have been indicated, typical for other kind of noise - such as caused by the measuring system and uncertainty characteristic for the corona AN - such as air humidity, rainfall intensity, masking of tonal components. Special attention has been paid to the partial uncertainty of random variable spectral structure of both the acoustic background noise and corona AN. On the other hand paper includes an assessment of uncertainty in 1/3 octave frequency bands and its influence on uncertainty of A-weighted level. The paper presents a practical approach of uncertainty calculation in determining long-term noise indicators for the AN using data gathered in the continuous monitoring system.