The Research Of Ocean Turbulence Detecting And Processing Based On Submerged Buoy

Turbulence is a high-frequency random movement， prevalent in the ocean. Thetemperature and salinity of the ocean will be changed by the phenomenon ofturbulence， and also the currents and marine climate. At the same time， the growthand distribution of the marine organism is impacted mostly， too. So， we shouldmaster and analysis the information of the ocean turbulence. It is important for us toextend the knowledge about the environment of the ocean and improve the predictionsystem on marine climate and disaster more reliable. At present， several countrieshave begun their research on the observing and sampling the turbulence， and theyalso has their own turbulence observing instrument. The acquisition method ofturbulence data has became the important part of the ocean observing technology.After the summarize of developing history and the experience with theturbulence measurement instrument research in other countries， this paper introducesa fixed-point measurement of ocean turbulence system (FMS-OUC)， which isdesigned by our own team. Each part is also described in the paper. We also designeda comparing experiment between a popular micro-scale profiler and our owninstrument. The results gave higher accuracy and precision in FMS-OUC.The turbulence measuring instrument will be interfered with many otherelements during the acquisition in the sea. In digital signal processing，time-frequency analysis by change the one-dimensional signal into two-dimensionaltime-frequency plane in order to effectively show the time and frequency ofnon-stationary signal changes in corresponding relationship of the local characteristicsof the signal， which reveals hidden deep message of the signals. Usually， we canuse short-time Fourier transform， Wigner-Ville Distribution and wavelet transform，etc. Although the traditional Fourier transform can not analysis the signal in timedomain， it is a classic method in the frequency domain of stationary signal analysis.The wavelet has a good performance in both time and frequency domain， it can givegood results about non-stationary signal. But， the environment of the ocean is socomplex that not any method can process the data effectively. Matching pursuitprocessing method can flexible decompose the original signal into waveforms basedon the characteristic of the signal. In the other way， we should build a dictionarywhich based on the turbulence in actually to make sure our method is reliable.This paper provides a good characteristic analysis for the turbulence andacceleration signals. Then， a new method of matching pursuit based on wavelet wasdeveloped. We also discuss the Fourier transform and wavelet， includes their usageand functions. For the turbulence data and acceleration data， different ways weretaken to process the signals. And in this paper， series of simulation experiments andoffshore trail provide actual data for us to get the real turbulence message， and the dissipation rating after dispose of vibration signal matches the Nasmyth spectrumwell.