The Research Of Transfer Path Analysis Methods Based On In-Situ Blocked Force And Multi-Subsystem Inverse Substructuring

The NVH performance of an automobile is a very important performance index,and it is also the main cause of various failures of automobiles.The transfer path analysis method(TPAM)is a good way to analyze the vibration and noise problems.However,the existing TPAMs are mainly based on experiments with high economy and time consumption,and are less combined with numerical simulation.It is more likely to be used for solving the specific problems of automobile NVH.It is rarely used for the optimization and forward development of automobile NVH.At the same time,the existing TPAMs mostly analyze NVH problems from the entire automobile perspective,and seldom conducts automobile NVH analysis and optimization from the subsystems’ perspective(such as excitation source characteristics and resilient mount characteristics,etc.);therefore,it has important application value to propose a TPAM that can analyze and predict the entire automobile NVH based on the subsystems’ characteristics.The research of this thesis is supported by the project “NVH performance improvement of driving axle by using TPA and system response methods” cooperated between Jilin University and an automobile company.According to the concept of TPA,the excitation source characterization methods and the dynamic substructuring /inverse substructuring theory,this paper proposes TPAMs based on in-situ blocked force and multi-subsystem inverse substructuring to analyze,predict and optimize the automobile vibration and noise during forward development,which will provide important theoretical basis and reference method.The main research work is summarized as follows:In order to validate the proposed TPAM,a motor-driven test rig system that simulates automobile engine excitation characteristics,resilient mount,frame and body is designed based on the transfer path analysis principle of the automobile vibration and noise.The advantages and disadvantages of the conventional TPAM(CTPAM)are verified by analyzing the vibration response of test rig system by using the matrx inversion TPA.Firstly,the TPA model is establisded for analyzing the vibration of the target response point of the test rig system and the tests of test rig system are carried out for the measurement of receiver substructure’ s FRFs and vibration response at various stationary rotation speed conditions;then it is analyzed that the effect of indicator number on the FRFs’ condition number and the results comparison of interface load identification accuracy by differenct truncation singular value decomposition methods(TSVDMs),and the threshhod value TSVDM is selected for the dynamic interface loads identification of test rig at ten stationary rotation speeds conditions.By the identificed interface loads,the paths contribution of target point is analyzed;after that,the FRFs of receiver substructure is calculated numerically by finite element method(FEM),which is compared with the experimental FRFs;finally all the analyzing results and characteristics of conventional TPA(CTPA).Although the CTPAM has a relatively high analysis accuracy,there is a huge defect,that is,the identified interface loads are not transplantable.When the receiverg substructure changes,the interface loads needs to be re-identified,and a lot of repeated tests are required,causing relatively high cost and long cycle of automobile NVH development.To solve the above-mentioned problems of CTPAM,based on the advantage that the in-situ blocked force can independently characterize the excitation source substructure and does not change with the receiver substructure,the in-situ blocked force TPA(In-situ bf TPA)is proposed,and it is extended to make it suitable for resilient coupling systems with numerical subsystems’ FRFs;and the proposed In-situ bf TPA is verified by the test rig system.Firstly,vibration response test and systematic FRFs measurement is performed under each operational condition;then the blocked forces at the three active ends of the test rig are identified based on the in-situ blocked force theory,by which the paths contribution is analyzed;secondly,the vibration responses of the response point are predicted based on the identified blocked force,and also compared with experimental value.In order to further verify the independence and transferability of the proposed In-situ bf TPA for different receiver substructures,the original structure test rig is structurally changed;then the response vibration of the changed test rig is analyzed by In-situ bf TPA,that is,the vibration tests of various operational conditions,system FRFs acquisition,blocked force identification,paths contribution analysis,and response point response prediction were carried out;then the blocked force of original test rig were compared with that of the changed test rig to validate the transferability of the identified blocked force of In-situ bf TPA to different receiver substructure.In order to indirectly prove the accuracy of the proposed In-situ bf TPA,the identified blocked force,the contribution of each transfer path and the predicted vibration response value of the target point obtained by In-situ bf TPA are compared with that of CTPA,of which the comparison result indirectly verifies the accuracy of the proposed In-situ bf TPA.However,in practical engineering,in order to fasten the NVH development of the whole automobile,the FRFs data of some substructures are obtained by numerical simulation analysis.In order to extend the proposed In-situ bf TPA to the resilient coupling system including the numerical simulation data of the substructure,the extended In-situ bf TPA is applied to the test rig;first,the FRFs of the excitation source substructure is tested by experiments,and the FRFs of the receiver substructure is obtained by numerical simulation;then,the dynamic stiffness of the resilient mount is experimentally measured;finally the vibration response of system target point is predicted.All the above analysis results verify the effectiveness and accuracy of the proposed In-situ bf TPA and the possibility to combine with numerically simulated FRFs data of receiver substructure for system response prediction.However,for a resilient coupling system composed of multi-subsystem,it is required to implement a large number of experimental tests to obtain the dynamic stiffness of the resilient mounts,which is a very complex,time-consuming and labor-intensive,and of which the accuracy is not high.Therefore,it is of great significance to propose one method to indirectly identify the mount stiffness,which can be conbined with TPA concept to solve the path contribution analysis of system with unknown resilient mount stiffness.To avoid much measurement cost and difficulty to obtain the unknown dynamic stiffness of resilient,the multi-susystem inverse substructuring TPA(MSISTPA)is proposed based on the inverse substructure theory of the multi-substructure system,by which the subsystems’ FRFs and dynamic stifness of resilient mount can calculated indirectly by systematic FRFs.In order to verify the accuracy and reliability of the proposed MSISTPA,the MSISTPA is numerically verified by using a multi-Dofs lumped parameter system;first,the multi-Dofs lumped parameter system is divided into different subsystems;secondly the FRFs of subsystems and dynamic stifness of resilient connector are identified by systematic FRFs based on multi-subsystem inverse substruturing theory,which is then compared with the actual value.The comparison results show that the proposed MSISTPA has high accuracy and better reliability;finally,the sensitivity analysis of each path’s force transmissibility and systematic FRFs to the stiffness of resilient connectors is analyzed.The proposed MSISTPA is also verified by the test rig system;first,the system simplification and substructure division of the test rig system are carried out,then the TPA model is established,and the vibration response test of each operational condition and the systematic FRFs acquisition are carried out;secondly,the FRFs of subsystems and the dynamic stiffness of the three resilient mounts are identified by the obtained systematic FRFs,which are all compared with the experimental values;then the interface force at the passive end of each resilient mount are identified based on the mount stiffness method and the identified mount dynamic stiffness;to verify the reliability of the MSISTPA,the identified interface force by MSISTPA are compared with the interface loads identified by the CTPAM,of which the comparison results show that the MSISTPA has high accuracy and reliability on path contribution analysis of target points;subsequently,the contribution of all paths under each operational condition is analyzed to provide a basis for the system vibration response control under a specific operational condition;finally,the force transmissibility of each path is analyzed based on the identified subsystems’ FRFs and the dynamic stiffness of the resilient mounts,which will provide a reference for the prediction and control of system response.All the above analysis results validate the reliability and accuracy of the proposed MSISTPA method to analyze and predict the system’s vibration and noise.And the MSISTPA can be completely applied to analyze and control the automotive NVH performance.