The problem of the structural noise of the submarine tail is studied.The propeller shaft coupling model of the submarine tail is selected as the research object.The submarine tail mass is taken as the constraint condition,and the radiated sound power level of the underwater submarine tail under the longitudinal and transverse excitation is taken as the optimization objective.A uniform experimental design is established with the thickness of the tail shell plate and the structural parameters of the T-section(panel width,web height,panel thickness,web thickness)as the design variables.The radial basis function(RBF)neural network is used to build a proxy model of the mapping relationship between design variables and optimization objectives,and particle swarm optimization(PSO)is used to optimize the submarine tail noise.The research shows that under longitudinal excitation and transverse excitation,the underwater radiated sound power synthesis levels of the submarine tail are reduced by 3.79 dB and 1.55 dB respectively,and the submarine tail mass is reduced by 3.424 t.Using RBF-PSO algorithm to optimize the low-frequency noise of submarine tail structure can obtain better effect.This work may provide a guidance for the optimization of submarine structure noise.
维普
万方数据