Deep learning method for analysis and segmentation of fatigue damage in X-ray computed tomography data for fiber-reinforced polymers

Helwing, Ramon; Hülsbusch, Daniel; Walther, Frank

doi:10.1016/j.compscitech.2022.109781

Cited by 17 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some enhancements would be possible using a dye penetrant to improve the local absorbance contrast if the damage cracks are connected to the surface [ 79 ]. Currently, some modern data analysis, such as deep learning or machine learning, has been proposed to identify defects in composite materials in X-ray images [ 80 , 81 , 82 ].…”

Section: Volume Non-destructive Testing Techniquesmentioning

confidence: 99%

Non-Destructive Health Monitoring of Structural Polymer Composites: Trends and Perspectives in the Digital Era

Chaki

Krawczak

2022

Materials

View full text Add to dashboard Cite

Non-destructive testing (NDT) techniques are usually used for the characterisation of defects arising in polymer composites during manufacturing or in-service use. However, each of these NDT techniques cannot always allow a full diagnosis of the material’s or component’s structural health. Thus, several techniques have to be combined in order to improve the diagnosis of the damaged state of composite structures and their evolution during the part’s life span. This opinion paper proposes a critical overview of the use and applicability of these NDT techniques for the detection and characterisation of damage to structural composite materials in view of in-service performance assessment and residual durability prognosis. It also addresses some current trends of structural health monitoring (SHM) of these materials, such as sensor–actuator embedding and NDT data fusion, and draws future perspectives on how composite SHM could evolve in the digital era, taking advantage of artificial intelligence, Internet of Things and big data to implement digital twins.

show abstract

Section: Volume Non-destructive Testing Techniquesmentioning

confidence: 99%

Non-Destructive Health Monitoring of Structural Polymer Composites: Trends and Perspectives in the Digital Era

Chaki

Krawczak

2022

Materials

View full text Add to dashboard Cite

show abstract

“…The vacuum-assisted resin transfer molding (VARTM) process has become a significant method for preparing composite wind turbine blades, naval hulls, and decks as well as bridge decks due to its cost-effectiveness, ease of operation process, and operation at room temperature. 1,2 However, the voids within composite components become one of the main defects when applying this technique. 3 There are several factors affecting the formation of voids in composites, including improper lay-up operations, resin volatiles released during curing, and moisture retention, among which air entrapment caused by microscopic inhomogeneous flow during mold filling is the main cause of voids generation.…”

Section: Introductionmentioning

confidence: 99%

“…The vacuum‐assisted resin transfer molding (VARTM) process has become a significant method for preparing composite wind turbine blades, naval hulls, and decks as well as bridge decks due to its cost‐effectiveness, ease of operation process, and operation at room temperature 1,2 . However, the voids within composite components become one of the main defects when applying this technique 3 .…”

Section: Introductionmentioning

confidence: 99%

Micro‐mesoscopic analysis of 3D void formation and evolution in CFRP composite using high‐resolution x‐ray microtomography

Peng,

Li,

Yang

2024

Polymer Composites

View full text Add to dashboard Cite

Void defect is prevalent during the composite manufacturing process and may reduce the mechanical properties of composite structures, while limited information about three‐dimensional void distribution and evolution can be found due to the difficulties in void observation. In this work, void defects at different positions were detected and analyzed using x‐ray microtomography (XRM), and the random forest algorithm was implemented for segmenting components of CFRP samples. Results showed a trend of gradual rising porosity in the axial direction, with an increase of ~250% of porosity at the outlet of resin injection compared to the one at the inlet during the composite manufacturing process. The voids mainly featured with microscale between carbon fibers, with less distribution between fiber bundles and resin‐rich zones. Void defects achieved aggregation at the middle position of manufactured laminate, which is probably related to pressure interception. In the axial direction, the proportion of large‐diameter voids gradually increased, and a large volume of connected voids appeared near the outlet. Due to the shear deformation of resin and fiber and the pressure gradient effect, most of the voids showed stripe shape. The spatial orientation of voids was predominantly along the direction of resin flow and fluctuated due to fiber curl.Highlights 3D void characteristics in CFRP were studied using the XRM technique. Void formation and the effects of fibers on void morphology were analyzed. Dynamic void transport mechanisms during CFRP manufacturing were studied. Location and morphology of voids are helpful for CFRP damage modeling. The effective partitioning of CFRP was achieved using a random forest algorithm.

show abstract

“…By combining micro-CT with image processing and computer vision technologies, new avenues have opened up for textile engineering, particularly in defect detection [6], fiber analysis [7], and material characterization [8]. Moreover, micro-CT provides digital twins of the scanned objects, enabling the usage of these 3D digital twins in modeling and simulation studies [9], [10].…”

Section: Introductionmentioning

confidence: 99%

Restoration of Textile Yarn Distorted Low-Resolution Micro Computed Tomography Cross Section Images: A MATLAB Restoration Algorithm

ABDELKADER,

PETRIK

2023

NANOCON Conference Proeedings

View full text Add to dashboard Cite

Textile yarn is a group of twisted fibers with diameters of a few micrometers, requiring a nano-resolution scanner to capture precise details of the single fibers perfectly to make a digital twine of the scanned yarn. Computed tomography (CT) technology can 3D digitally scan the sample and achieve a digital twin. The fine fibers' diameter requires a nano-CT to achieve a high-resolution yarn's digital twin; nano-CTs are more expensive than micro-CTs and require about eight times the scanning time compared to micro-CTs, which means more computational power to reconstruct and analyze the scanned objects. This paper introduces a systematic MATLAB algorithm to regenerate distorted yarn's micro-CT low-resolution cross-section images. The algorithm segments the distorted images' fibers, identifies them, and regenerates the clean, highresolution fibers. The algorithm performance is compared to the optical microscopic cross-section image measurements using ImageJ. The results revealed that before processing, the mean fiber diameter measured 9.60 ± 0.78 µm, while post-processing it measured 10.33 ± 0.49 µm. Notably, the algorithm effectively decreased the dispersion of fiber diameters around the mean by 40%, maintaining a diameter close to the design diameter of the fibers of 10 µm.

show abstract

Deep learning method for analysis and segmentation of fatigue damage in X-ray computed tomography data for fiber-reinforced polymers

Cited by 17 publications

References 32 publications

Non-Destructive Health Monitoring of Structural Polymer Composites: Trends and Perspectives in the Digital Era

Non-Destructive Health Monitoring of Structural Polymer Composites: Trends and Perspectives in the Digital Era

Micro‐mesoscopic analysis of 3D void formation and evolution in CFRP composite using high‐resolution x‐ray microtomography

Restoration of Textile Yarn Distorted Low-Resolution Micro Computed Tomography Cross Section Images: A MATLAB Restoration Algorithm

Contact Info

Product

Resources

About