Comparison between Finite Element and Hybrid Finite Element Results to Test Data for the Vibration of a Production Car Body

The Hybrid Finite Element Analysis (HFEA) method is based on combining conventional Finite Element Analysis (FEA) with analytical solutions and energy methods for mid-frequency computations. The method is appropriate for computing the vibration of structures which are comprised by stiff load bearing components and flexible panels attached to them; and for considering structure-borne loadings with the excitations applied on the load bearing members. In such situations, the difficulty in using conventional FEA at higher frequencies originates from requiring a very large number of elements in order to capture the flexible wavelength of the panel members which are present in a structure. In the HFEA the conventional FEA model is modified by de-activating the bending behavior of the flexible panels in the FEA computations and introducing instead a large number of dynamic impedance elements for representing the omitted bending behavior of the panels. The excitation is considered to be applied on the conventional FEA model and the vibration analysis is conducted using the FEA. The power flow through the dynamic impedance elements is computed, and in this work it is used to assess the vibration of the flexible panels. Successful comparisons between test data and HFEA results have been presented in the past for car body structures and for a rotorcraft application. In this paper a three-way comparison is presented for the vibration of a production vehicle body in the frequency range 200Hz - 1,000Hz. Six different excitation locations on the stiff parts of the structure are utilized (one at a time); for each excitation the mobility of five reference body points on the load bearing members, and the mobility of seven flexible vehicle panels is measured and the measurements are compared with both conventional FEA results and with HFEA computations. The conventional FEA model which is used in this work for the respective computations is a production FEA vehicle model with sufficient refinement in order to be valid up to 1,000Hz based on a requirement of exhibiting at least six elements per wavelength at the highest frequency of interest. The FEA model had been validated in the past through comparison with modal test data in low frequencies. Discussion about the development of the HFEA model and the correlation of both numerical solutions to the test data are presented in this paper.