The Research On Bandgap Characteristic Of Elastic Metamaterial With Multilayered Honeycomb Structure And Its Bandgap Optimization

Elastic metamaterials are artificially periodical composite materials with elastic wave band gaps,within which the elastic wave propagation is totally prohibited.The bandgap characteristic provides potential applications in vibration attenuation and noise reduction for engineering application,therefore,it is significant to design EMs with low-frequency bandgaps.In this thesis,an elastic metamaterial with a multilayered honeycomb structure is initially proposed,and its bandgap characteristics are analyzed.In order to broaden bandgaps in the lowfrequency range,several optimization approaches,including the finite element method combined with genetic algorithm(FEM-GA)and the genetic algorithm with Kriging-based surrogate optimization method(Kriging-GA)are proposed to solve the optimization problem for maximizing the bandgap considering the structural lightweight requirements.The main contents of the thesis are as follow:(1)In order to obtain broadband attenuation,an elastic metamaterial with a multilayered honeycomb structure is proposed.In conjunction with the vibration modes at bandgap edges and the analytical model,the formation mechanism of the bandgaps is analyzed,and the effects of various geometry parameters of the bandgap characteristic are discussed.(2)The optimal problem of the elastic metamaterials with a multilayered honeycomb structure for maximizing the sum of relative bandgap widths with a given structural mass constraint is established.Then FEM-GA was employed for solving the optimization problem,in which the finite element method is used to evaluate the fitness function in the optimization framework,while the genetic algorithm is adopted to find the near optimal solution.The results indicate that the optimized solutions can improve the sum of relative bandgap widths significantly However,the computational effort may become prohibitive since the tremendous amount of evaluating fitness function through the finite element method.(3)In order to tackle the bottleneck of FEM-GA,an optimization approach integrating Kriging surrogate model and genetic algorithm(Kriging-GA)is proposed.Firstly,an initial Kriging model is conducted by using data with known objective values,and the Kriging model is updated based on an adaptive infill criterion.Then this trained model is used to replace the FEM calculation in the FEM-GA,and the genetic algorithm embedded with Kriging model is used to search for the optimized design variables.The optimized results shows that the KrigingGA optimization approach is capable of improving the computational efficiency while ensuring optimization effectiveness,which provides widespread application prospect.