Skip to Main Content
 

Global Search Box

 
 
 
 

Files

ETD Abstract Container

Abstract Header

Finite Element Model Correlation with Experiment Focusing on Descriptions of Clamped Boundary Condition and Damping

Abstract Details

2020, PhD, University of Cincinnati, Engineering and Applied Science: Mechanical Engineering.
A two-step approach is developed to build an FEM model of a clamped structure focusing on the boundary stiffness and damping definitions. The approach utilizes FE model correlation with experimentally obtained modal parameter estimates to calibrate the model. Model calibration is first carried out in a free-free state to update the geometry, mass and material properties of the structure. The calibrated free-free model is then updated in two stages to include the boundary stiffness and damping for a clamped state. Stiffness is addressed first followed by damping. The stiffness is defined in terms of contact stiffness definitions in terms of normal and tangential stiffness at the boundary. Contact stiffness is determined by matching analytical and measured natural frequencies of the clamped built-up structure. Calibration is carried out in a mode-wise manner and the system response is calculated in the frequency domain in bands centered around each mode of interest. In the second step, the damping property of the structure is identified based on responses at resonance. The significance of spatial distribution of damping is studied by first developing an approach to include accurate spatial distribution in models based on intuitive knowledge of the location of damping and then using it to compare against models built with traditional damping models such as Rayleigh and modal damping. Accurate representation of spatial distribution of damping in FE models was observed to be not very important for lightly damped structures. The stiffness and damping modeling approaches developed were combined and demonstrated on a clamped steel beam. The calibrated model is then validated by demonstrating its ability to make accurate strain predictions to arbitrary load cases. Random broadband and banded chirp loads were both used to compare the system responses from simulation and testing. Simulated FRF from calibrated system and the force spectra of interest are used to obtain the frequency domain response predictions. To obtain the time domain response an IFFT was used. Good agreement was observed between the response predicted in this manner and the one measured using the experimental setup.
Jay Kim, Ph.D. (Committee Chair)
Randall Allemang, Ph.D. (Committee Member)
Allyn Phillips, Ph.D. (Committee Member)
Yongfeng Xu, Ph.D. (Committee Member)
160 p.

Recommended Citations

Citations

  • Jayakumar, V. (2020). Finite Element Model Correlation with Experiment Focusing on Descriptions of Clamped Boundary Condition and Damping [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1592171762884289

    APA Style (7th edition)

  • Jayakumar, Vignesh. Finite Element Model Correlation with Experiment Focusing on Descriptions of Clamped Boundary Condition and Damping. 2020. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1592171762884289.

    MLA Style (8th edition)

  • Jayakumar, Vignesh. "Finite Element Model Correlation with Experiment Focusing on Descriptions of Clamped Boundary Condition and Damping." Doctoral dissertation, University of Cincinnati, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1592171762884289

    Chicago Manual of Style (17th edition)