Dataset for scanning electron microscopy based local fiber volume fraction analysis of non-crimp fabric glass fiber reinforced composites

Lars P Mikkelsen; Søren Fæster; Vedrana A Dahl

doi:10.1016/j.dib.2023.109058

Dataset for scanning electron microscopy based local fiber volume fraction analysis of non-crimp fabric glass fiber reinforced composites

Data Brief. 2023 Mar 31:48:109058. doi: 10.1016/j.dib.2023.109058. eCollection 2023 Jun.

Authors

Lars P Mikkelsen¹, Søren Fæster¹, Vedrana A Dahl²

Affiliations

¹ DTU Wind, Technical University of Denmark, DK-4000 Roskilde, Denmark.
² DTU Compute, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark.

Abstract

This dataset includes four large field-of-view scanning electron microscopy (SEM) images together with associated Matlab scripts aimed for the analysis used in the joint publication. Each of the four stitched images is generated from a large number (between 15500 and 24500) high-resolution (195nm/pixel) scans, which have been stitched into four images stored as tiff-files. The images show the cross-section of fiber bundles in composite laminate and are well-suited for local fiber volume determination. The image resolution corresponds to between 600 and 2000 pixels covering each fiber. The imaged samples are from composite laminates with an overall fiber volume fraction in the range of 55% to 60%. The local fiber volume fraction is found both for the full cross-section, as an average fiber volume fraction over the individual bundles, and as a local fiber volume fraction found in a moving averaging box with a size corresponding to 5×5 fiber diameter (80×80 µm²) areas.

Keywords: Bundle segmentation; Fiber volume fraction; SEM; Wind turbine blades.