Design Optimization Of Acoustic Structural Coupled Systems Considering Random Parameters

Most researches on traditional acoustic structural coupled systems are based on deterministic methods.However,in practical engineering,due to assembly and measurement errors and other factors,there will be randomness in the material parameters,sizes and other aspects of the acoustic structural coupled systems.Normally,single random parameter may have little effect on the system,but when multiple random parameters act simultaneously,the response of the coupled system tends to cause a large deviation.Therefore,the influence of random parameters must be considered in the analysis and optimization of acoustic structural coupled systems.In order to solve the above problems,this thesis uses pseudo-excitation method,reliability method and stochastic perturbation method to study the influence of random factors on response analysis of acoustic structural coupled systems.At the same time,for these three methods,the corresponding stochastic optimization models of the coupled system are established and the effective optimization algorithms are proposed.The main works of this thesis are as follows:1.Analysis and design optimization of acoustic structural coupled systems under stationary random excitation.Firstly,the finite element model of the acoustic structural coupled systems is established.Furtherly,the random response analysis of the acoustic structural coupled systems under single-point random excitation and multi-point random excitation is performed based on the pseudo-excitation method;then,the size optimization and topology optimization models of the acoustic structural coupled systems are established,and the sensitivity analysis formulas of the sound pressure power spectral density to the design variables are studied.Finally,taking the average sound pressure power spectral density in the bandwidth as the objective function and the structural mass or volume as the constraint,the optimization solution is carried out.Numerical examples results show that the objective function is significantly reduced and the effectiveness of the optimization method is demonstrated.2.The size optimization design considering the reliability index of the acoustic structural coupled systems,based on the Kriging surrogate model.Firstly,the Kriging surrogate model is fitted by adopting the optimization sampling algorithm based on the feasible domain to establish the relationship between the structure size and the performance function.Then,the mass of the structure is used as the objective function,and the reliability index as the constraint function.The inner layer uses the Monte-Carlo method to calculate the reliability of the performance function,and the outer layer uses the genetic algorithm to solve external layer optimization.Numerical examples show that compared with the deterministic optimization,the mass of the structure in the reliability optimization result increases slightly,the design of the latter is more reliable and safe than that of the former.3.Analysis and topology optimization design of random acoustic structural coupled systems based on stochastic perturbation method.First,the stochastic finite element model of the acoustic structural coupled systems is established,and the random analysis based on the stochastic perturbation method is carried out.Then,the robust topology optimization model of the stochastic acoustic structural coupled systems is established,and the corresponding sensitivity analysis formulas are derived.The method of moving asymptotes is used to complete the optimization solution.Numerical examples show that the results of the stochastic perturbation method are consistent with those of the Monte-Carlo method,and the sound pressure expectation and variance in the results of robust topology optimization are smaller than those of the traditional deterministic optimization.