Gradual Acoustic Fault Prediction of Submarine

Abnormal vibration noise created by equipment during operation will degrade the acoustic stealth performance of submarine, and increase the possibility of the boat exposing to the underwater acoustic threat in hostil e water field. This paper aims to obtain a rate of mechanical noise sources＂ contribution to the overall underwater noise of submarine via multi-input/multi-output (MIMO) model. In order to find out abnormal conditi ions of submarine, the time series of noise source＇s contribution rate were analyzed for determining the latent acoustic fault. Following the track of acoustic fault together with regulating the calculation interval o of noise sources＂ contribution, the fault time would be linearly predicated according to the tendency of contribution time series. The feasibility of the proposed approach has been validated by an experimental model.