An Internet of Things Based Crack Monitoring Approach Using Nondestructive Evaluation Data

Monitoring and quantifying crack initiation and growth are of primary importance both for material performance evaluation and design and for structural damage assessment. While several sensing and evaluation methods related to cracking have been proposed, recent demands for real-time assessment have created the need to connect data acquisition with rapid extraction of information that can be leveraged in both diagnostics and prognostics. This investigation presents a novel approach to leveraging nondestructive evaluation (NDE) datasets in an internet of things (IoT) framework, which is shown to be capable of providing nearly real-time diagnostics for cracking, while creating also the framework to apply prognostics methods. To demonstrate this approach, compacttension specimens of an aluminum alloy were used in laboratory experiments in accordance with ASTM standards. Acoustic emission NDE datasets were acquired and used to produce information that was processed using an in-house-built IoT system capable of edge and cloud computing. The main innovation of this approach is that a combination of IoT hardware and software proves advantageous in implementing a data structure that can then be used in machine learning operations that are suitable for detecting crack initiation.