PSI - Issue 66

Hendrik Baarssen et al. / Procedia Structural Integrity 66 (2024) 305–312 Author name / Structural Integrity Procedia 00 (2025) 000–000

306 2

Nomenclature k

Number of parameters

v

CMOD Compliance, measured as displacement between the measurement points

A ij

Displacement field matrix Thickness of C(T)-specimen Modulus of Elasticity

B E F N P S U

Chow coefficient

Number of observations

Load

Residuals

Normalized compliance parameter

W Effective width of C(T)-specimen, measured as distance between the center of the pinholes and back of the specimen CMOD Crack Mouth Opening Displacement C(T) Compact Tension DCPD Direct Current Potential Drop DIC Digital Image Correlation FCGR Fatigue Crack Growth Rate 1. Introduction In some structural applications such as railway, aerospace, and bridge design, the design against fatigue failure can be based on a damage tolerant design philosophy, hence quantifying the Fatigue Crack Growth Rate (FCGR) is necessary to determine the reliability and remaining safe lifetime of structural elements in which a fatigue crack nucleated. Therefore, the FCGR of a given material is usually experimentally determined following existing standards such as ASTM E647 (ASTM International, 2000). In any case, a standard specimen is submitted to cyclic loading, often but not necessarily of constant amplitude and mean stress, and the crack size is measured as a function of the number of applied cycles. Several sensors and methods exist to measure the FCGR. This is most often done using crack gauges, strain gauges, and clip-on gauges. Whilst the first sensor directly measures the location of the crack tip discretely and with a resolution depending on the wire pitch, with strain gauges and clip-on gauges it is possible to obtain a continuous measurement by making use of back-face and crack mount compliance relationships, (ASTM International, 2000). Techniques based on potential drop can provide accurate measurements but require relatively expensive equipment and an electrically isolated experimental setup. In recent years, Digital Image Correlation (DIC) has become increasingly popular for a variety of measurements including fatigue crack growth. Choi and Shah (1997) demonstrated the potential of DIC-based crack detection by measuring the displacement field of concrete subjected to compression. Ruocci et al. (2016) expanded the methodology by applying a moving average filter to reduce the noise on the displacement field. The strain field is probed along a line across which cracks are expected. These are identified by peaks in the strain field, i.e. when the strain exceeds the mean strain plus one standard deviation, which are calculated over the line. This is then repeated for each measuring line creating a 2D surface of the specimen, allowing to detect cracks independent of the direction. Later, Gehri et al., (2020) developed a script to determine the width of the crack using the principal strain field, and the crack width using the deformation field. Since the direction and the magnitude of the principal strains are known, the cracks can be detected in all directions. The threshold is based on the yield strain of the material and the measured noise level. Gehri et al. (2022) improved the script using a method proposed by Canny (1986) for edge detection thus enhancing the crack kinematic measurements. The method proposed by Canny is commonly used to detect edges in image processing techniques and has been applied to detect the maximum principal strain, which allows the detection of cracks once it exceed the assumed threshold value.