PSI - Issue 39

Available online at www.sciencedirect.com Available online at www.sciencedirect.com ScienceDirect StructuralIntegrity Procedia 00 (2019) 000–000

www.elsevier.com/locate/procedia

ScienceDirect

Procedia Structural Integrity 39 (2022) 608–623

7th International Conference on Crack Paths A Review of Application of Data Science Tools in Crack Identification and Localization Nabam Teyi a *, Sandeep Singh a a Department of Mechanical Engineering,North Eastern Regional Institute of Science and Technology, Nirjuli, Arunachal Pradesh 791109, India Abstract A rotating shaft supported by bearings over which a disc is attached concentric to it, as a single unit is called a rotor, which forms the basis of functions of many machine systems. A rotor’s structural integrity and adherence to the central axis during its static/rest state or dynamic/rotating state, is a major concern for engineers. Ideally there must be zero deflection of the disc or shaft from the axis of rotation at all circumstances, which do not happen in reality. A beam is a structural element that resists lateral loads (an element designed to carry primarily axial load would be a strut or column) and it mostly bends to deflect. The loads applied to the beam generate reaction forces at its support points. A beam is also a subject of induced flaws due to constant resistance to bending and deflection. Neither a rotor shaft nor a beam is free from some defects. Structural defect due to presence of a small opening like a slit cut on its surface called as a crack is ubiquitous. Identification of crack presence and then its location therefore has become a continued pursuit of maintenance engineers. However, these structural members may not be disengaged from their machineries for crack identification and localization, and so, crack analysis is performed when the machines are operating. Some data acquisition devices are put around these shafts and beams and the output signals are analysed. The output signal is the dataset generated by the system. These dataset are needed to be tested with different data science tools for the crack analysis. Also, cracked rotor and beam systems offer a wide possibility of analytical, numerical, computational modelling and experimental investigation. A lot of authors have done enough works in crack analysis in transverse, axial or slant directions, on single, double or multiple cracks. This paper presents a review of such chosen literatures that reported application of some data science tools in crack analysis in shaft and beams. © 2021 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of CP 2021 – Guest Editors Keywords: Crack identification; Crack localization; Aritifical Neural Network; Data Science; Machine Learning; Artificial Intelligence ep Singh a vices are put © 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of CP 2021 – Guest Editors

* Corresponding author. Tel: +91 8974393401 E-mail address: nbtnerist@gmail.com; nbt@nerist.ac.in

2452-3216© 2021 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of CP 2021 – Guest Editors

2452-3216 © 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of CP 2021 – Guest Editors 10.1016/j.prostr.2022.03.134