| Warship Engine is a complex power equipment used on warships, and assembles many systems including mechanical, hydraulic and electrical, etc. Compared with ordinary air engine, its working environment is relatively simple and crude. It is affected by the spray of salt and heat of the sea and is prone to breakdown. As a result, the research of the warship engine fault prediction technology, which can be used to predict the early failure of the warship engine, to ensure the safety of the warship engine, is of great significance to protect the warship's combat power.In this paper, the key component of a certain type of warship is used as the research object, whose method and application technology of fault prediction are studied. Firstly, the fault types and fault state evolution characteristics of the warship engine bearing are deeply analyzed. Secondly, the vibration detection method is used, in a certain type of real warship engine test platform of the long-term test, to obtain the bearing vibration acceleration signal. The original data using CEEMD method for noise reduction processing reduce the interference in the original data. Once again, according to the characteristics of the signal after the noise reduction, the parameters that can be used to characterize the health status of the bearing should be obtained. A feature extraction method based on wavelet packet energy is proposed, which is used as the characteristic information of the prediction of the running state of the engine bearing. Finally, the fault prediction model based on grey theory, the fault prediction model based on grey neural network and the fault prediction model based on RVM are established by using the energy information of the wavelet packet, and are also used to predict the health status of the warship engine bearing. The research shows that the fault prediction method based on RVM is the best, and has a good application prospect in the warship engine bearing fault prediction. |
PDF Full Text Request