@article{LI2025, 
author = {Hefan LI and Guanjun ZHANG and Yuzhao KE},
title = {Low-noise optimization design for underwater series-connected multi-sphere composite shell structure},
year = {2025},
journal = {Chinese Journal of Ship Research},
volume = {20},
number = {3},
pages = {118-124},
keywords = {shell (structure), spherical shell, composite shell, radial basis function (RBF) neural networks, particle swarm optimization (PSO), noise, design optimization, noise optimization},
url = {https://www.sciopen.com/article/10.19693/j.issn.1673-3185.03619},
doi = {10.19693/j.issn.1673-3185.03619},
abstract = {ObjectiveTo maximizing the working performance of the series-connected multi-sphere composite shell structure, an optimization design study has been conducted to enhance its acoustic and vibration performance. MethodsFirst, a finite element model was developed to calculate and analyze the underwater acoustic radiation characteristics. Then, using the model's mass as the constraint condition and the radiation noise under vertical excitation as the optimization objective, a uniform experimental design was developed using the geometric parameters of the connecting structure as the design variables. The radial basis function (RBF) neural network was used to establish a multidimensional mapping model between the design variables and the optimization objective. The particle swarm optimization (PSO) algorithm was employed to optimize the underwater radiation noise of the model, and the results were verified through the finite element method. Finally, underwater acoustic radiation experiments were carried out, and the experimental test values were compared with the simulation values to validate the accuracy of the simulation results.ResultsThe results show that after optimization, the total underwater radiation sound power level of the series-connected multi-sphere composite shell under vertical excitation was reduced by 2.92 dB, while the mass was decreased by 0.061 t.ConclusionThe research provides new insights into the optimization of low-noise structural design methods.}
}