Journal Published Online: 09 May 2018
Volume 7, Issue 2

Development of an Optimization Tool for Calibrating Crack Growth Material Models



Crack growth models are typically calibrated using experimental data prior to conducting damage tolerance analyses of aircraft structures. This process can be trivial for crack growth models that only have a few fitting parameters. However, other models, such as the FASTRAN crack growth equation with the analytical crack-closure model, include several model parameters that make calibration challenging. To simplify the calibration process and improve the accuracy of crack growth simulations, an automated crack growth model optimization tool was developed. This tool also aims at improving the robustness of crack growth models by reducing their dependency on loading spectra and specimen geometries. To achieve this, numerical optimization is used to minimize the discrepancy between experimental data and analytical crack growth. To demonstrate this approach, material model parameters were calibrated for P-3/CP-140 aircraft applications using the FASTRAN crack growth equation with the analytical crack-closure model. The optimization tool was found to be very effective at fitting crack growth simulation results to experimental data by changing the values of selected parameters. It was, however, observed that the automated calibration process could find multiple sets of FASTRAN parameter values that provide equivalent correlations with experimental data. In an attempt to develop a more robust material model, multiple test configurations with different geometries and loading spectra were used to determine an optimal trade-off model that maximizes the correlation for multiple test configurations. The conducted tests demonstrated that this approach is viable and could be used to improve the robustness of crack growth simulations.

Author Information

Bombardier, Yan
Aerospace Portfolio, National Research Council Canada, Ottawa, ON, Canada
Pages: 16
Price: $25.00
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Stock #: MPC20170086
ISSN: 2379-1365
DOI: 10.1520/MPC20170086