Session: 12-16-03: Drucker Medal Symposium

Paper Number: 100137

100137 - Direct Heterogeneous Material Property Determination Using Full-Field Experimental Measurements 

A direct approach is described to determine the elastic modulus distribution in a nominally heterogenous material. For the case of linear elastic response, the matrix equations are in the inverted form [A] {E} = {R}, where the [A] matrix components are known functions of measured strains and axial positions, {R} represents components that are known functions of specified body forces, applied loads and reactions and {E} represents components of the unknown elastic properties at discrete locations within the specimen.

Focusing on nominally uniaxial specimens undergoing tensile/compression loading, {E} represents values of the elastic modulus at node locations along the specimen length.  Since digital image correlation (DIC) methods are envisioned to obtain the full-field deformation data that is input into the inverted finite element model for modulus determination, results from 1D studies are performed to determine the data density, N_d, required for accurate reconstruction of the modulus distribution. In this regard, the relationship between an appropriate DIC subset size, DIC strain window size, FE element size and the required data density is discussed and results shown for selected 1D simulations. In these studies, the effect of random variability in the measured strains is shown to demonstrate the accuracy and noise sensitivity of the methodology. In addition to simulation-based studies, the 1-D method is used with uniaxial strain measurements in a bone specimen undergoing compressive loading to determine the spatially varying elastic modulus. To demonstrate the potential of the method, our 1-D model predictions are directly compared to an independent set of experimental modulus measurements. Results demonstrate that the predicted spatial variations in modulus are in very good agreement with the independent experimental measurements at each load level of interest along the length of the bone specimen. Since extension of the method to non-linear response is of general interest, an incremental linear approach is demonstrated for monotonic loading to successfully extend the approach and predict the spatial distribution in secant modulus, E_S(x), at multiple load levels, developing a multi-linear approximation at each location. Using this approach, the local stress-strain response of the heterogenous bone material at four distinct locations can be determined. 

For applications where two-dimensional deformations are measured and nominally plane stress conditions exist (e.g., free surfaces), recent developments are described using basic theoretical constructs, resulting in the same general computational form with matrices in the inverted form [A] {E} = {R}. In this application, {E} can have multiple parameters at each location (for example , Young’s modulus and Poisson’s ratio). Preliminary results for the two-dimensional studies are presented, along with a brief discussion of future extensions to three-dimensions and dynamic loading.

Presenting Author: Michael Sutton Univ. of South Carolina

Presenting Author Biography: Michael Albert Sutton was born in the small town, Carmi, Illinois, on February 5, 1950. He received his Ph.D. in 1981 from the Department of Theoretical and Applied Mechanics at the University of Illinois under the direction of Prof. Charles E. Taylor (1923-2017). After graduation, Prof. Sutton joined the Department of Mechanical Engineering at the University of South Carolina in 1982. Upon promotion to full professor in 1992, he was awarded a Carolina Distinguished Professorship and retained the Chair for the remainder of his career. He co-founded Correlated Solutions Inc. (CSI) in 1997 to transfer the digital image correlation (DIC) technology he pioneered to industry and government laboratories and has served as. Chief Science Officer since its founding. Today, CSI is the only US-based DIC R&amp;D center and provider of DIC measurement systems. <br/> Prof. Sutton has received numerous honors for his contributions in conceptualization, development, application and world-wide technology transfer of digital image correlation software and hardware. He was designated a Fellow of both the Society for Experimental Mechanics (SEM) in 2000 and ASME in 2004. He is past President of SEM (2001-02) and founding President for the International Digital Image Correlation Society (2014-2018). He received an honorary doctorate degree from Ecole Polytechnique-Cachan/Saclay in 2011, the first individual ever selected based on contributions in experimental mechanics. He received the Murray Medal and Lectureship from SEM in 2013. He was elected to the US National Academy of Engineering in 2020 and the Slovenian National Academy of Engineering in 2021. In 2022, he received the SES Medal from the Society for Engineering Sciences (2022) and the Timoshenko Medal from ASME. <br/> Prof. Sutton’s current areas of research interest include extension of StereoDIC for deformation measurements in civil infrastructure, including (a) railroad rails, (b) prestressed concrete beams, (c) concrete and masonry walls and (d) use of drones for visual measurements on large civil structures. Prof. Sutton has also been active in measuring the response of composite tows during manufacturing of aerostructures and civilian infrastructure, including measurement of traction-separation laws for manufacturing simulations.<br/> Prof. Sutton married Elizabeth Ann Severns on September 8, 1973 in Carmi, Illinois. They have two daughters, Michelle S. Spigner and Elizabeth S. Gosnell, and five grandchildren. They continue to live near Columbia, SC on a 12.5 acre farm/nature preserve, which they opened to the public during the pandemic, providing a place for home-bound families to get away and enjoy nature.

Authors:

Michael Sutton Univ. of South Carolina
Sreehari Rajan-Kattil University of South Carolina
Subramani Sockalingam University of South Carolina