Transfer Path Analysis Of Vibration And Noise In Ships Based On Neural Networks

The performance of acoustic stealth mainly depends on the level of noise generated by the ship,and controlling noise is a key factor to consider in ship acoustic design.During ship cruising,the main radiated noise of the ship is mechanical noise.In order to achieve targeted noise reduction in the future,it is necessary to find the main transfer path of vibration noise.Due to the complexity of the transfer process of ship vibration and noise,existing methods cannot accurately and quickly find the vibration transfer path.Therefore,it is of great significance to study the process of radiating noise from ships from mechanical equipment to the shell and ultimately to the underwater acoustic evaluation point,in order to identify the main transfer path of noise.This article conducts research on the response prediction,contribution analysis,and vibration transfer path analysis of ship mechanical structure vibration noise.It systematically analyzes the process of ship cabin noise propagating from mechanical equipment to the shell and ultimately radiating into the underwater acoustic field,and proposes certain improvements to the existing theory of working condition transfer path.The main work completed in this article is as follows:1.In response to the issue of the inability to determine whether important transfer paths are missing in the transfer path of the working conditions,a heavily coherent analysis based on biased dryness is proposed to verify the completeness of the input and output signals and avoid missing important transfer paths.And it was integrated into the data preprocessing process of the neural network-based condition transfer path analysis method,improving the processing process of the neural network-based condition transfer path analysis method.2.In response to the problem of low estimation accuracy of the transfer function in traditional transfer path analysis methods,a neural network-based method for working condition transfer path analysis is proposed.By adjusting the structural parameters of the neural network,a neural network suitable for analyzing the vibration transfer characteristics of ships is obtained.Afterwards,the adjusted neural network was used to compare and analyze the advantages and disadvantages of the traditional transfer path analysis method and the method proposed in this paper.Simulation calculations showed that the neural network method can increase the ability of the transfer path analysis method to resist crosstalk and noise interference,and improve the accuracy of obtaining the transfer function.In addition,heavy coherence analysis is introduced to avoid the omission of important transfer paths.3.In response to the analysis of the vibration and noise transfer characteristics of complex structures in naval vessels,a multi-level operating condition transfer path analysis model for naval vessel cabin sections based on neural networks was established based on the analysis of operating condition transfer paths.According to the propagation order of mechanical equipment noise,the ship cabin is divided into multiple subsystems.The data obtained from cabin model tests and real ship cabin tests are used to analyze the transfer characteristics between measurement points of each subsystem.Finally,the main transfer path of vibration noise is analyzed,and the model is used to predict vibration noise,verifying the effectiveness of the method.