Restricted by observation instruments and methods, the measured turbulence signals in the open ocean are inevitably affected by noise. To maximum eliminate the noise caused by the vibration of the instruments and improve the accuracy of the measured turbulence signals, an improved turbulence signal denoising algorithm based on least mean square (LMS) adaptive filter is proposed in this paper. The key point of the improved algorithm is to obtain the optimal weight value by the adaptive adjustment and remove the vibration noise from the measured shear signals in frequency domain. Through taking the Nasmyth theoretical spectrum as the desired signal to update the weight value, we can obtain the optimal estimation of the observation spectrum, especially at the shedding frequency. Sea data obtained from a moored turbulence measuring instrument (MTMI) deployed in the South China Sea is used to verify the feasibility and validity of the improved algorithm. The results show that the corrected spectra agree well with the Nasmyth theoretical spectra and the calculated dissipation rates of turbulent kinetic energy (TKE) drop near an order of magnitude compared with the raw measured data, which indicate that the improved turbulence denoising algorithm can effectively eliminate the vibration noise and provide accurate data for studying the characteristics of turbulence mixing.

Included in

Engineering Commons