Studies Of Structural Vibration And Noise Control Based On Multi-objective Optimization Method

Structural vibration and noise control widely exist in aerospace,nuclear industry,and shipbuilding industry.Structural vibration control problems often include multiple objectives and high-dimensional design variables.Usually,these problems have conflicting objectives,such as lightweight design and vibration reduction performance,the design of dynamic vibration absorbers for vibration control and lightweight design simultaneously,etc.Moreover,due to broadband excitation,low-frequency vibration and noise control,and nonlinearity,many difficulties are brought to structural dynamics analysis and design.Therefore,to meet the industry’s new demands,it is important to study a new method for structural vibration and noise control.This thesis starts with the vibration and noise control issues of elastic beams.First,the mass and radiated sound power of elastic beams are taken as two objectives.The thickness distribution and the distribution of the functionally graded materials are considered as design variables.The transfer matrix method is used to analyze the vibroacoustic properties of the non-uniform beams.Many numerical results are presented to show the importance of using the multi-objective method.In addition,the nonlinear dynamics of the acoustic black hole beam,the coupling effect of the system with both acoustic black hole and nonlinear energy sink,the band gap properties and sound insulation design of beams with multiple acoustic black holes are also studied.These studies proposes new prospects of acoustic black holes.For the structural optimization algorithm,based on the cell mapping method,this thesis proposes a new hybrid method of particle swarm optimization and cell mapping.In the early stage of the iteration,The particle swarm algorithm is used to generate an initial set of random points as the estimate of the Pareto solution.The cell mapping method is then applied to the covering set with a search and recovery procedure.The hybrid method is not only more accurate but also more reliable in terms of its ability to solve complex problems with either continuous or disjointed Pareto sets than the cell mapping method applied separately.This method is used to study the optimal design of beams for noise reduction.To improve the computational efficiency,this thesis combines a global database and local database to establish surrogate models,which can be switched adaptively in the iterative process.This algorithm can obtain the solution set with fewer function evaluations and is applied to the sound insulation design of periodic acoustic black hole structures and the optimal design of two-dimensional metamaterial plates.In recent years,metamaterials have received widespread attention.This thesis studies the one-dimensional metamaterial based on acoustic black holes and halfsinusoidal corrugates.The relationship between lightweight design,structural strength constraints,and vibration performance is investigated.In order to improve the structural strength,this paper proposes a periodic stiffened panel.The finite element method is used to analyze the band gap mechanism.Two panels are made by using 3D printing,and the vibration transmission is measured to verify the band gap.In addition,a metamaterial plate based on acoustic black holes is proposed,and its vibration reduction mechanism is revealed through modal analysis.Results demonstrate the rich localresonance modal behavior of the acoustic black holes.The main contents of this thesis are presented above.We start with the vibration control problem of elastic beams and then study the dynamic analysis and design of one-dimensional metamaterial beams and two-dimensional metamaterial plates.Two multi-objective optimization algorithms are proposed for structural design.In the study of metamaterials,both numerical values and experiments verify the efficiency and effectiveness of the proposed optimization algorithms and vibration control ability of metamaterials,respectively.The research work in this paper will provide efficient analysis and optimization strategies for vibration analysis and optimization design of engineering structures.