Research On Uncertainty Theory And Application For Automobile High-frequency Noise

The performance of automotive NVH(Noise,Vibration,Harshness)can be divided into low-frequency performance,middle-frequency performance and high-frequency performance,among which the performance of high-frequency NVH has the most significant impact on the sound quality of vehicle.At present,the development of high-frequency NVH performance of whole vehicles is mainly solved by SEA(Statistical Energy Analysis)method.The material,structure and performance parameters in the SEA model are all deterministic parameters.However,in practical engineering problems,the excitation and environmental conditions are constantly changing,and the manufacturing,assembly and measurement errors cannot be eliminated,so the uncertainty of automobile,especially its acoustic package system,is inevitable.These uncertain factors interact and couple with each other,resulting in a large deviation between the actual performance of the automotive high-frequency NVH and the design performance,which can not be ignored on the performance of large-scale products.Aiming at the existing problems in the development process of the Automotive high-frequency NVH performance,uncertainty theory and algorithm are introduced in this paper to systematically study the performance development theory and application of the Automotive high-frequency NVH under uncertain conditions,and strive to make exploratory progress in the performance uncertainty analysis theory and application of Automotive high-frequency NVH.First of all,on the theoretical level,the stochastic statistical energy method and interval statistical energy method are proposed for the uncertainty analysis of automotive acoustic-structural coupling model.Secondly,at the application level,in order to overcome the interval expansion phenomenon caused by the uncertainty of the design variables,an interval perturbation analysis model considering the coupling relation between uncertain variables is proposed.Thirdly,in order to predict the Automotive high-frequency noise response deviation caused by material parameter uncertainty propagation,the uncertainty of Automotive high-frequency NVH performance is analyzed based on the interval model.Finally,an efficient interval robustness optimization method is introduced to solve the optimization problem of high-frequency NVH performance of the whole vehicle considering uncertainties.In this case,the research works in this thesis are as follows:(1)For the case that the distribution information of the SEA parameters in the automobile acoustic-structural coupling model is difficult to obtain,the uncertainty theory is introduced into the analysis process of SEA,and the stochastic statistical energy method and interval statistical energy method for the uncertainty analysis of the automobile acoustic-structural coupling model are proposed referring to the stochastic finite element and interval finite element theory.The material parameters of structure and acoustic cavity are selected as the uncertainty parameters,which not only make the description of the uncertainty parameters more direct and easy at the initial design stage,but also avoid the influence of the interaction between SEA parameters on the uncertainty analysis results.Based on the statistical energy equation,the partial derivative equations of the loss factor matrix with respect to material parameters of the structure and acoustic cavity are derived by analytical method,and the quantitative relationship between the subsystem energy fluctuation and the material parameter uncertainty is established,which has higher precision and wider applicability than numerical method.Taking the simple flat-cube acoustic cavity coupling model and the car engine compartment-firewall-interior acoustic cavity model as examples,the accuracy and universality of the proposed methods are verified by comparing the calculation results of the Monte-Carlo method,and the high-frquency response characteristics of the system under uncertain conditions are predicted successfully.(2)A new relevant interval perturbation method for the uncertain analysis of high-frequency sound insulation performance of the key automotive assembly is presented.Considering the linear inequality constraint relation during the uncertain parameters,the sensitivity ranking mechanism of uncertain parameters is introduced.This mechanism overcomes the limitation of the independence of uncertain parameters in the traditional interval model.The accurate calculation method is researched and presented which could be effective to suppress the interval expansion of the system response in the uncertain analysis.Taking the automotive firewall as an example,the SEA model of the assembly is established based on the statistical energy method,and the sound absorption and insulation performance of acoustic package and through-hole parts are obtained through experiments,and the SEA model is adjusted based on the test data.The relevant interval perturbation method is used to analyze the sound insulation performance of the firewall assembly;the numerical example shows that the model and method developed in this work can effectively solve the problem of coupling during uncertain parameters.Taking the mass of the inner dashboard as the objective function and taking the sound insulation performance of the firewall assembly as the constraint condition,the parameters of the inner dashboard are optimized based on the multi-island genetic algorithm to achieve the goal of lightening the inner dashboard while still improving the robustness of the sound insulation performance of the firewall assembly.(3)For the uncertainty of porous material parameters,an interval uncertain analysis method for the high-frequency noise performance of the automobile is presented.The absorption coefficient and transmission loss of the acoustic package are obtained through tests,and a statistical energy analysis model of the whole vehicle is established.The acoustic loads of the vehicle are tested and the interior noise of cabin is analyzed under certain working conditions.Uncertain parameters are introduced to describe the acoustic package system of the firewall,and the interior noise response is predicted through the interval perturbation method,and the robustness of the system under the influence of uncertain parameters is analyzed.(4)A highly efficient robustness optimization design method for improving the performance of the automotive acoustic package is presented.The full vehicle model is established based on the statistical energy analysis method and the accuracy of the model is verified through acoustic transfer function testing.The parameters affecting the sound absorption and insulation performance of the key acoustic packaging parts are selected as the uncertain parameters and their sensitivity is analyzed.The possibility degree method of interval numbers is introduced to convert the uncertainty constraints to deterministic constraints,and then an efficient decoupling method is introduced to convert the two-layer nesting robustness optimization model to a single-layer optimization model,and the efficient robust design of the automotive acoustic packaging is realized.The acoustic packaging parts of a sport utility vehicle are analyzed and optimized using the proposed efficiently robustness optimization method,and the robustness of the system is greatly improved.