N221-007 TITLE: Data-Driven Physics-Based Modeling Tools to Determine Effective Mechanical Properties of As-Built Composite Structures

OUSD (R&E) MODERNIZATION PRIORITY: Artificial Intelligence (AI)/Machine Learning (ML)

TECHNOLOGY AREA(S): Air Platforms;Materials / Processes

OBJECTIVE: Develop a software toolkit to automate the generation of nonlinear, anisotropic mechanical properties for as-built composite structures, including the effects of defects, to accelerate finite element (FE) analysis for fleet repairs and aircraft production non-conformal dispositions.

DESCRIPTION: Advanced rotors for vertical lift aircraft and wings on many U.S. Navy fixed-wing aircraft are complex assemblies made primarily from thermoset composites (e.g., IM7/977-3). Full-scale fatigue tests are frequently required to qualify and certify these critical safety items for a calculated number of flight hours. The selected part chosen for testing may have deliberate seeded flaws and/or severe manufacturing defects to capture the worst damage/condition expected in service. After these weakened parts survive the full-scale fatigue tests, applied knockdown factors further reduce the risk of fatigue failure. Even though the strength safety margin for a given part could be sufficiently high, when service damage occurs, engineers have very tight repair limits, and few options, due to the fatigue life constraint. Local stress distributions, and in-situ mechanical properties of the composite parts, have a significant influence on fatigue life and residual strength, and are very complex to predict, especially for the thick (0.5 in./1.27 cm or greater) laminate composites.

A potential remedy to establish additional cost-effective repair options is to implement a data-driven, physics-based, modeling approach by analyzing the parts in the as-built condition with their own unique configuration, including manufacturing defects and in-service damages. Examples of manufacturing defects are wrinkles, marcels, foreign object debris, porosities/voids, and delamination. In-service damages could include impact, maintenance induced, and heat or ballistic damage.

In addition to an accurate FE mesh representation of the as-built component, the other crucial analysis requirement is the assignment of accurate in-situ (nonlinear) mechanical properties to the FEs. Typical mechanical properties for laminate composite FE analyses (FEAs) use linear orthotropic values based on coupon testing (versus as-built structures). As a result, strain gauge values monitored during full-scale tests can differ substantially from FEA results. These differences between strain gauge results and strain/stress analysis predictions deserve scrutiny when considering repair options. Innovative advancements in computerized tomography (CT) scan image processing coupled with advanced micro-meso-macro mechanics modeling can be utilized to yield not only more representative anisotropic mechanical properties, but also a more accurate stress/residual strength analysis of the real structures.

The Navy seeks to develop a software toolkit that can automate the process to generate in-situ, nonlinear, anisotropic effective mechanical properties using CT scans of as-built composite parts. The critical size and boundary conditions of the representative volume element (RVE) must be consistent with the material system�s inhomogeneity, scan resolution, and fidelity of the intended FE mesh. The scan resolution should be sufficiently high enough to capture the appropriate length scale(s) associated with material system components (e.g., ply thickness/orientation, fiber path/bundle/volume, fiber/resin, and adhesive interfaces) and manufacturing defects (e.g., porosities/voids, wrinkles, delamination, and fiber waviness). The most critical defects include combinations of wrinkles, porosities/voids, and resin-rich or adhesive-rich zones, which should be captured by the model with an effective relationship to the FE mesh and intended analysis. The proposed toolkit must also account for material degradation due to repeated loadings and Hot/Wet (H/W) operating environments. The generated quasi-static and dynamic-effective mechanical properties (stiffness, strength, and strain energy release rate) must be compatible with different 2-D and 3-D FE types including shell/plate and tetra-/hexahedral elements. Since the data volume of the CT scans could be very huge (larger than one terabyte) for a full-scale component, speed and accuracy issues relating to data acquisition, image processing, and data storage and retrieval must also be addressed, including the use of machine learning (ML) and computer vision techniques.

PHASE I: Demonstrate technical feasibility of the proposed concept to develop a computationally efficient, multiscale, physics-based, modeling toolkit coupled with CT-scanned data, ML, and computer vision techniques to generate in-situ, quasi-static, and dynamic effective mechanical properties (stiffness, strength, and strain energy release rate) for as-built, thick laminate composite structures, including effects of defects, repeated loadings, and expected H/W operating environments. Demonstrate the proposed workflow to auto-populate the input data for different 2-D and 3-D FE meshes, including various element sizes and types to support progressive damage analysis of thick laminate composite structures. Develop a verification and validation (V & V) test plan for the proposed concept, including, at a minimum, the use of Digital Image Correlation (DIC).

PHASE II: Perform CT scan of test coupons/components representative of a structural component with manufacturing defects (e.g., L-shape). Develop algorithms for fast CT image processing, automated feature extraction, and identification/classification with ML techniques, and data storage and retrieval. Demonstrate the generation of a localized FE mesh from CT scan data capturing ply orientations and manufacturing defects. Demonstrate the integrated process utilizing the developed multiscale, physics-based, modeling toolkit and CT-scanned data to predict the in-situ, quasi-static, and dynamic effective mechanical properties (stiffness, strength, and strain energy release rate) for a structural representative thick laminate composite test component including effects of defects and operating environments. Demonstrate the auto-populated input data functionality for different 2-D and 3-D meshes. Conduct testing in accordance with the V & V test plan developed in Phase I to correlate with the predicted results.

PHASE III DUAL USE APPLICATIONS: Finalize the prototype modeling-toolkit and ensure usability for the end user. Perform final testing to demonstrate the toolkit�s ability to support analysis of a fleet repair or solve a production issue on a large-scale and relevant platform part.

Commercial aviation uses similar structures and has a similar need for more capable analysis toolkits to analyze repairs and production issues. This capability might also find use in the wind turbine industry, as the blades are large composite structures.

REFERENCES:

1. Avril, S., Bonnet, M., Bretelle, A. S., Grédiac, M., Hild, F., Ienny, P., Latourte, F., Lemosse, D., Pagano, S., Pagnacco, E., & Pierron, F. (2008, July 25). Overview of identification methods of mechanical parameters based on full-field measurements. Experimental Mechanics, 48(4), 381. https://doi.org/10.1007/s11340-008-9148-y.
2. Rahmani, B., Villemure, I., & Levesque, M. (2014). Regularized virtual fields method for mechanical properties identification of composite materials. Computer Methods in Applied Mechanics and Engineering, 278, 543-566. https://doi.org/10.1016/j.cma.2014.05.010.
3. Straumit. I., Lomov, S., & Wevers, M. (2015). Quantification of the internal structure and automatic generation of voxel models of textile composites from X-ray computed tomography data. Composites Part A: Applied Science and Manufacturing, 69, 150�158. https://doi.org/10.1016/j.compositesa.2014.11.016.
4. Makeev, A., Seon, G., Nikishkov, Y., Nguyen. D., Mathews, P., & Robeson, M. (2016, May 17-19). Analysis methods improving confidence in material qualification for laminated composites [Paper presentation]. Proceedings of the American Helicopter Society 72nd Annual Forum, West Palm Beach, FL, United States. https://www.researchgate.net/publication/303562174_Analysis_Methods_Improving_Confidence_in_Material_Qualification_for_Laminated_Composites.
5. Nikishkov, Y., Seon, G., Makeev, A., & Shonkwiler, B. (2016, March 15). In-situ measurements of fracture toughness properties in composite laminates. Materials and Design, 94, 303-313. https://doi.org/10.1016/j.matdes.2016.01.008.
6. Seon, G., Makeev, A., Cline, J., & Shonkwiler, B. (2015, September 29). Assessing 3D shear stress-strain properties of composites using Digital Image Correlation and finite element analysis based optimization. Composites Science and Technology, 117, 371-378. https://doi.org/10.1016/j.compscitech.2015.07.011.
7. Lambert, J., Chambers, A. R., Sinclair, I., & Spearing, S. M. (2012, January 18). 3D damage characterisation and the role of voids in the fatigue of wind turbine blade materials. Composites Science and Technology, 72(2), 337-343. https://doi.org/10.1016/j.compscitech.2011.11.023.
8. Kim, H. J., & Swan, C. C. (2003). Algorithms for automated meshing and unit cell analysis of periodic composites with hierarchical tri-quadratic tetrahedral elements. International Journal for Numerical Methods in Engineering, 58(11), 1683-1711. https://doi.org/10.1002/nme.828.
9. Raju, B., Hiremath, S. R., & Mahapatra, D. R. (2018). A review of micromechanics based models for effective elastic properties of reinforced polymer matrix composites. Composite Structures, 204, 607-619. https://doi.org/10.1016/j.compstruct.2018.07.125.
10. Bargmann, S., Klusemann, B., Markmann, J., Schnabel, J. E., Schneider, K., Soyarslan, C., & Wilmers, J. (2018). Generation of 3D representative volume elements for heterogeneous materials: A review. Progress in Materials Science, 96, 322-384. https://doi.org/10.1016/j.pmatsci.2018.02.003.
11. Naresh, K., Khan, K. A., Umer, R., & Cantwell, W. J. (2020). The use of X-ray computed tomography for design and process modeling of aerospace composites: a review. Materials & Design, 190, 108553. https://doi.org/10.1016/j.matdes.2020.108553.
12. Dutra, T. A., Ferreira, R. T. L., Resende, H. B., Guimarães, A., & Guedes, J. M. (2020). A complete implementation methodology for Asymptotic Homogenization using a finite element commercial software: preprocessing and postprocessing. Composite Structures, 245, 112305. https://doi.org/10.1016/j.compstruct.2020.112305.
13. Nsengiyumva, W., Zhong, S., Lin, J., Zhang, Q., Zhong, J., & Huang, Y. (2021). Advances, limitations and prospects of nondestructive testing and evaluation of thick composites and sandwich structures: A state-of-the-art review. Composite Structures, 256, 112951. https://doi.org/10.1016/j.compstruct.2020.112951.
14. Thor, M., Sause, M. G., & Hinterhölzl, R. M. (2020). Mechanisms of origin and classification of out-of-plane fiber waviness in composite materials�A review. Journal of Composites Science, 4(3), 130. https://doi.org/10.3390/jcs4030130.
15. Zambal, S., Eitzinger, C., Clarke, M., Klintworth, J., & Mechin, P. Y. (2018, July). A digital twin for composite parts manufacturing: Effects of defects analysis based on manufacturing data. In 2018 IEEE 16th International Conference on Industrial Informatics (INDIN) (pp. 803-808). IEEE. https://doi.org/10.1109/INDIN.2018.8472014.
16. Badran, A., Marshall, D., Legault, Z., Makovetsky, R., Provencher, B., Piché, N., & Marsh, M. (2020, September 8). Automated segmentation of computed tomography images of fiber-reinforced composites by deep learning. Journal of Materials Science 55, 16273�16289. https://doi.org/10.1007/s10853-020-05148-7.
17. Fritz, N. K., Kopp, R., Nason, A. K., Ni, X., Lee, J., Stein, I. Y., Kalfon-Cohen, E., Sinclair, I., Spearing, S. M., Camanho, P. P., & Wardle, B. L. (2020). New interlaminar features and void distributions in advanced aerospace-grade composites revealed via automated algorithms using micro-computed tomography. Composites Science and Technology, 193, 108132. https://doi.org/10.1016/j.compscitech.2020.108132.

KEYWORDS: Composites; Finite Element Analysis; Damage Progression; Material Characterization; Manufacturing Defects; Composite Repairs; Computed Tomography