Abstract
A novel noise reduction method and corresponding technique are presented for improving turbulence measurements with acoustic Doppler velocimeters (ADVs) commonly used in field studies of coastal and nearshore regions, rivers, lakes, and estuaries. This bifrequency method is based on the decorrelation of the random and statistically independent Doppler noise terms contained in the Doppler signals at two frequencies. It is shown through experiments in an oscillating grid turbulence (OGT) tank producing diffusive isotropic turbulence that a shift in carrier frequency of less than 10% is sufficient to increase the resolved frequency range by a decade in the turbulent velocity spectra. Over this spectral range, the slope of the velocity spectra agrees well with the universal inertial range value of −5/3. The limit due to spatial averaging effects over the sample volume can be determined from the abrupt deviation of the spectral slope from the −5/3 value. As a result, the relative error of the turbulent intensity estimate and the turbulent kinetic energy (TKE) dissipation rate, measured by two different methods, does not exceed 10% in the case of isotropic turbulence. Furthermore, the bifrequency method allows accurate estimates of the turbulent microscales as shown by the good agreement of the ratio between the Taylor and Kolmogorov microscales and an Re1/4t power law. Compared to previous Doppler noise reduction methods (Garbini et al.), an increase in time resolution by a factor of 4 is achieved. The proposed method also avoids the loss of TKE energy contained in isotropic flow structures of size equal to and smaller than the sample volume. Different from Doppler noise methods proposed by Hurther and Lemmin and Blanckaert and Lemmin, this method does not require additional hardware components, electronic circuitry, or sensors because the redundant instantaneous velocity field information is captured with the same transducer. The required shift in carrier frequency is small enough for the bifrequency method to be easily implemented in commercial ADVs.
Text area detection in handwritten documents scanned for further processing
