In situ health monitoring of multiscale structures and its instantaneous verification using mechanoluminescence and dual machine learning

Seong Yeon Ahn; Suman Timilsina; Ho Geun Shin; Jeong Heon Lee; Seong-Hoon Kim; Kee-Sun Sohn; Yong Nam Kwon; Kwang Ho Lee; Ji Sik Kim

doi:10.1016/j.isci.2022.105758

In situ health monitoring of multiscale structures and its instantaneous verification using mechanoluminescence and dual machine learning

iScience. 2022 Dec 7;26(1):105758. doi: 10.1016/j.isci.2022.105758. eCollection 2023 Jan 20.

Authors

Seong Yeon Ahn¹, Suman Timilsina¹, Ho Geun Shin², Jeong Heon Lee³, Seong-Hoon Kim³, Kee-Sun Sohn⁴, Yong Nam Kwon⁵, Kwang Ho Lee⁶, Ji Sik Kim¹

Affiliations

¹ Kyungpook National University, School of Nano & Advanced Materials Engineering, Kyeongbuk 37224, Republic of Korea.
² Kyungpook National University, Department of Advanced Science and Technology Convergence, Kyeongbuk 37224, Republic of Korea.
³ Kyungpook National University, Department of Software, Kyeongbuk 37224, Republic of Korea.
⁴ Sejong University, Nanotechnology and Advanced Materials Engineering, 209 Neungdong ro, Gwangjin-gu, Seoul 143-747, Republic of Korea.
⁵ Korea Institute of Materials Science, Department of Digital Platform, Changwon 51508, Republic of Korea.
⁶ Kyungpook National University, Department of Automotive Engineering, Kyeongbuk 37224, Republic of Korea.

Abstract

Extensive changes in the legal, commercial and technical requirements in engineering fields have necessitated automated real-time structural health monitoring (SHM) and instantaneous verification. An integrated system with mechanoluminescence (ML) and dual artificial intelligence (AI) modules with subsidiary finite element method (FEM) simulation is designed for in situ SHM and instantaneous verification. The ML module detects the exact position of a crack tip and evaluates the significance of existing cracks with a plastic stress-intensity factor (PSIF; $K_{P}$ ). ML fields and their corresponding $K_{p}^{M L}$ values are referenced and verified using the FEM simulation and bidirectional generative adversarial network (GAN). Well-trained forward and backward GANs create fake FEM and ML images that appear authentic to observers; a convolutional neural network is used to postulate precise PSIFs from fake images. Finally, the reliability of the proposed system to satisfy existing commercial requirements is validated in terms of tension, compact tension, AI, and instrumentation.

Keywords: Machine learning; Mechanical Phenomenon; Optical property.