Research Of Uncertain Analysis And Optimization For The Structural-acoustic Coupled System Based On The Surrogate Model Technique

With the increasing demands on sound quality,the noise control of the enclosed cavity such as the car passenger compartments and the airplane cabins is becoming a challenging problem encountered by many engineers.The noise in the enclosed cavity is significantly affected by the vibrating structure,acoustic cavity and coupled solid-fluid interface.Traditional numerical methods for prediction and optimization of frequency response of the structural-acoustic coupled system are conducted under the assumption that the physical properties and boundary condtions.However,a structural-acoustic coupled system often contains many uncertain factors,such as the model inaccuracies,the unpredictable loading conditions,and the aggressive environment factors.In most case,the uncertain levels of uncertain parameters are small,but the uncertain parameters are various,and the couple effect of each uncertain parameters on the frequency response of the structural-acoustic coupled system is nonnegligible.As the uncertain parameters has some influence on the property of the structural-acoustic coupled system,if the structural-acoustic coupled system is designed by deterministic optimization method,the results may be beyond the constraint conditions when considering the uncertainties.Thus,it is necessary to study the prediction method and optimization method considering the uncertainties in the structural-acoustic coupled system.This dissertation aims to contribute some useful studies and trials on the uncertain analysis methods and uncertain optimization methods,especially on the uncertain methods based on surrogate modeling technique.The uncertain analysis methods are proposed to predict the frequency response of the structural-acostic system with pure random uncertainties,pure interval uncertainties and hybrid uncertainties.Combining the uncertain methods and optimization methods,this thesis conducted a systematical study on the uncertain optimization of structural-acoustic coupled system.The main work in the dissertation includes:Based on the non-intrusive polynomial chaos method,the dissertation presents a surrogate model for analysisig the structural-acoustic coupled system with random uncertainties.First,based on series expansion to model the relationship between input and output quantities.Secondly,obtain the probalistic information on the frequense response using the feature of the polynomial chaos expansion.Addtionally,for the structural-acoustic coupled system with random uncertainties containing hybrid distribution feature,we introduce a polynomial chaos method with hybrid bases,and derive the equations for calculating the probabilistic information on the frequency response.Finally,the effect of uncertainty level,confidence interval and hybrid bases on the accuracy and effectiveness of the proposed method is studied.It is found that polynomial chaos method can be efficiently and effectively to predict the probabilist information of frequency response of the structural-acoustic coupled system.For the structural-acoustic coupled system with interval uncertaintiy,the dissertation compares the accuracy of three different interval response surface method to calculate the intervals of fresponse response.This dissertation presents a novel subinterval response surface method for the structural-acoustic coupled system with relative large uncertainties.Firstly,carry out the sensitivity analysis of interval parameters.Secondly,determining the number of subinterval of each interval parameters based on the senistivtity proportion of interval parameters.Thirdly,obtain the intervals of frequency response of each subinterval using interval response surface method.Finally,obtain the intervals of frequency response of the whole interval.Numerical results show that subinterval interval response surface improve the accuracy of traditional interval response surface with a little increase of computational burden.The dissertation proposes an uncertain analysis method named Polynomial Chaos Response Surface method(PCRSM).Based on the PCRSM,the surrogate model is bulit for uncertainty propagation in structural-acoustic coupled system containing hybrid uncertainties.The fundamental idea of the proposed method is that the frequency response of the structural-acoustic coupled system with random and interval variables can be approximated by sums of polynomials of random and interval variables.The process of approximating the structural-acoustic frequency response manily consists of two steps.Firstly,the vector of interval variables is regarded as a constant vector.Based on the polynomial chaos expansion method,the coefficents can be obtained.Secondly,once the random bases is chosen,the coefficients depend only on the vector of interval variables.Then,the bounds of the response statistics can be obtained by combinging interval response surface method and MCM.The PCRSM does not require efforts to modify model equations due to its non-intrusive characteristic.Thus,the PCRSM shows superiority in application in large and complicated structural-acoustic coupled system.Furthermore,the PCRSM is also able to predict the response of the structural-acoustic coupled system with only random or only interval uncertainty.The dissertation presents a novel optimization scheme under random uncertainty and a new optimization scheme under random and interval uncertainty for conducting reliability and robust optimization of structural-acoustic coupled system.Based on the surrogate model obtained by polynomial chaos method and PCRSM,using the HMSPO method to optimize the mass of the structural-acoustic coupled system with random uncertainty and hybrid uncertainty,respectively.It is found that the results obtained by the proposed methods can effectively reduce the mass of the structural-acoustic coupled system with good performance of frequency response.An experiment is carried out to study the effect of thickness of plate and environment temperature on the frequency response of a 3D structural-acoustic coupled system,and the effectiveness of the surrogate model built by interval response surface method is validated.