In this paper a detailed comparison is performed between an Acceleration-based transfer path analysis (TPA) and a Strain-based TPA with as goal to evaluate and judge the accuracy of the obtained TPA model and estimated body forces. The overall TPA result, the target prediction quality, is high for both the Acceleration and Strain-based approach. However, an evaluation of the body-forces that each contribute to the total target response shows large differences between the loads obtained with the different force identification approaches. Overall the body-forces estimated with the Strain-based TPA are significantly lower, especially for loads that are expected to be small, such as the X- and Z-direction load for a lateral suspension link connection. An example of overestimated Z-direction loads from the Acceleration-based TPA is shown, where two close-by paths have an identical high force amplitude, while the these forces are in counter-phase. This results in a minimal combined contribution to the total target response. To have an objective indication which force identification approach enables the most accurate load identification, a measurement is performed in which 2 known forces are applied to the Trimmed Body, to close-by connection points in the same direction. Using the 2 approaches, these forces are then also estimated using either the Accelerationor Strain-indicators. Comparison with the actually applied force shows that the Strainbased approach results in the most accurate force-identification. Additionally, since only 2 forces are applied to the Trimmed Body, all other forces have to be zero. The Strain-based TPA model results in 2 high level forces while the other forces are of significant lower level. The Acceleration-based TPA results in multiple high level forces - also the surrounding ('zero-level') forces are estimated at a similar level as the applied loads. This approach is not able to accurately separate the forces that are identified in this TPA model. When accurate loads are required for application to a CAE body model, use of strainbased load identification proves to have a significantly higher accuracy compared to acceleration-based load identification in case of a structure with strong cross-coupling between the input points.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen


    Exportieren, teilen und zitieren



    Titel :

    Accuracy of inverse load identification techniques for transfer path analysis


    Beteiligte:
    Sachse, Daniel (Autor:in) / Geluk, Theo (Autor:in) / Wayenberge, Tom van (Autor:in)


    Erscheinungsdatum :

    2013


    Format / Umfang :

    11 Seiten, 8 Bilder, 6 Quellen


    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Datenträger


    Sprache :

    Englisch




    Accuracy of inverse load identification techniques for transfer path analysis

    Sachse,D. / Geluk,T. / Van Wayenberge,T. et al. | Kraftfahrwesen | 2013



    Development of transfer path analysis for road noise using parametric load model

    Shiraishi, Y. / Wada, M. / Ishizuka, M. et al. | British Library Conference Proceedings | 2012



    Practical Dynamic Load Identification of Vehicle Systems with Quasi-Static Analysis and Inverse Problem

    Federation Internationale des Societes d'Ingenieurs des Techniques de l'Automobile | British Library Conference Proceedings | 2002