Research On Analysis Methods For Mechanical Vibration Signal Of Underwater Vehicle

The underwater vehicle generally means underwater weapon system such as submarine, torpedo and so on. The level of mechanical noise of underwater vehicles is one of main index of strategy and tactic in acoustic warfare, which restricts battle effectiveness and survivability. In this dissertation, we combined various types of major components of the underwater vehicle prototype, installed the acceleration sensors on the outer surface of the prototype, and sample the signals. In accordance with the results of the sample processing, we select the combination prototype with a set of the lowest mechanical vibration.In this dissertation,bearing vibration signals of underwater vehicle prototype was gathered in the different states, the vibration signal of roller bearings is very complex, including not only the information of bearings self but also information of other related parts and structures. It is difficult to find fault fore-and-aft transformation, if we only use time-domain or frequency-domain means to analyze these signals. If time-frequency method is used at the same time, the diagnostic veracity and reliability will be greatly improved. So we study Time-Frequency analysis method. Short Time Fourier Transform, Wavelet Transform, Wigner-Ville distributing and Choi-Williams distributing that belong to Time-Frequency analysis are used to analyze bearing vibration signals of the underwater vehicle prototype. We select the most suitable bearing with low-noise operation for the underwater vehicle by analyzing the fault signals of the actual bearing and contrasting the keenness of non-stationary signals separately using Short Time Fourier Transform, Wavelet Transform, Wigner-Ville distributing and Choi-Williams distributing.