Parametric identification for material of viscoelastic SHPB from wave propagation data incorporating geometrical effects

2015 
Abstract In the application of Split Hokinson Pressure Bar (SHPB) techniques, test bars made of viscoelastic materials are frequently used for their low impedance to test soft materials. Wave propagations in such bars are susceptible to dispersions caused by the viscoelastic effects of the bar material along with the lateral inertia of the bar particularly when large diameters are used. Thus, the time shifting of the strain signals from the recording gauges on the bars to the bar-specimen interface requires corrections based on both the material properties and the diameter of the bar. Most of the methods available in the open literature, for viscoelastic material identification using wave propagation tests, are based on one-dimensional wave propagation theory in which signal dispersions due to the lateral inertia are ignored and hence are appropriate for the bars with small diameters. A method of parametric identification is proposed in the current study, using the frequency domain information, for the viscoelastic properties of a relatively large diameter bar made from polymethyl methacrylate (PMMA). A three-dimensional wave propagation analysis is employed using recorded strain data at a single location on the PMMA bar. It is found that the proposed parametric identification procedure, compared with the results of a non-parametric identification, helps to rectify the uncertainties in the identified material characteristics in the form of a highly oscillatory trend caused by the noise in the recorded signals and by the data reduction procedures. Strain signals predicted using the identified material parameters by the proposed method correlate well with the measured experimental strains, indicating accuracy in the identified parameters and the presented procedure.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    23
    References
    17
    Citations
    NaN
    KQI
    []