Study On The Vibration Signal Analysis Of The Unmanned Undersea Vehicle Based On The Wavelet

Concealment is of prime importance for an unmanned undersea vehicle's operational performance. The key to improve the concealment of unmanned undersea vehicle (UUV) is the reduction of noises, while the mechanical vibration is the main resource of the noises, so it has practical significance to study on the vibration signals of UUV and then to reduce the noises through reducing vibration.The vibration signals of UUV are very complex and contain a large amount of information about moving parts and the work of structure, so it is difficult to determine whether the status of their work is normal if we analysis the vibration signals of UUV only from the time domain or from frequency domain. There is no doubt that it will improve the accuracy and reliability of diagnosis if we combine the wavelet analysis, which is good at time domain analysis, with conventional detection method, which features frequency domain analysis. This thesis made a processing project based on wavelet analysis for the vibration signals of UUV, by collecting vibration signals through doing vibration experiments on several kinds of stern section of UUV,made comparison among conventional signal processing method, continuous wavelet transform method and wavelet packet analysis method, proposed a diagnosis method not only based on wavelet analysis but also applied to vibration signals of UUV and we named the new method "Trend– Total Energy Level" diagnosis. We processed and analyzed a large number of experimental data in this way, ultimately achieve the experiment aims to found the vibration sources of the faults, optimized the shell structure and reduced the noises of UUV. From what we have done in this thesis we can see: wavelet analysis technology provides a powerful tool to diagnose the vibration signals of UUV.