Title
Challenges in the phase identification of steels using unsupervised clustering of nanoindentation data
Conference Dates
October 2 – 7, 2022
Abstract
Cluster analysis tools are used in data interpretation to separate information without the bias of a user. In the current study we investigate two techniques, the elbow method and K-means clustering to achieve a phase classification for a dual phase (DP) and a high strength low alloy (HSLA) steel by using hardness and reduced Young’s modulus from nanoindentation tests as input variables. For the DP steel the contrast in hardness of the two phases ferrite and martensite is high, while for the HSLA steel the hardness contrast between ferrite and bainite is small, as seen from the corresponding load-displacement curves (Fig. 1).
Please click Download on the upper right corner to see the full abstract.
Recommended Citation
Gerhard Dehm, Robin Jentner, James Best, and Christoph Kirchlechner, "Challenges in the phase identification of steels using unsupervised clustering of nanoindentation data" in "Nanomechanical Testing in Materials Research and Development VIII", Sandra Korte-Kerzel, RWTH Aachen University, Germany Eds, ECI Symposium Series, (2022). https://dc.engconfintl.org/nanomechtest_viii/82