288 / 2021-04-19 11:23:28
Error analysis on inertial parameter identification of a rigid body based on mass line method
Rigid body inertial parameters; mass line method; error analysis; correlation; matrix condition number
Draft Accepted
Peibao Wu / Tsinghua University
Zhichao Hou / Tsinghua University;State Key Laboratory of Automotive Safety and Energy
Rongkang Luo / Tsinghua University
The inertial parameters of a rigid body are critical for subsequent dynamics analysis. The mass line method is a typical frequency-domain method that has been successfully applied in quite a few engineering fields. An error analysis method of the mass line method is presented. The errors are divided into the measured Frequency response functions (FRFs) noises and algorithm procedures errors. For the measured FRFs under each excitation, the correlation between the measured accelerations of all response points and the accelerations fitted by the rigid body motion equation in the selected frequency band is used to evaluate the measurement noises. The relative error of estimated rigid-body accelerations at the reference origin, mass center position and inertia are derived respectively. And the condition number of the measurement points coordinate matrix and reference origin acceleration matrix are employed to represent the sensitivity of the matrix calculation to the measurement noises, which mainly depends on the selection of the response and excitation points. Then the influence of different numbers and combinations of response and excitation points on the mass line method’s errors are discussed. Moreover, a criterion for response and excitation points selection is proposed. Finally, an engine is taken as an example to validate the validity of the error analysis method. The study is of great value to improve the inertial parameter identification accuracy.
Important Date
  • Conference Date

    Nov 01

    2022

    to

    Nov 03

    2022

  • Oct 30 2022

    Draft paper submission deadline

  • Nov 09 2022

    Registration deadline

Sponsored By
Qingdao University of Technology