Hybrid physics-based AI-enabled crack-length estimation from AE signal signatures 20-000001027
Abstract
Research problem Acoustic emission is well established as a nondestructive evaluation for monitoring the structural health by listening to the "pops" or hits generated by the energy released by an incremental crack growth. Existing AE equipment identified such hits by setting a threshold on the recorded structural wave signals. The threshold is set high-enough to discard "noise" signals originating from internal rubbing of crack faying surfaces or from the testing equipment and retain only the AE related to crack growth. Although being "the most crucial step in AE monitoring", the setting of the correct AE threshold remains an art form strongly dependent of the subjective intuition of experienced AE technicians. Most importantly, the current AE practice does not possess anearly warning capability because AE hit rates accelerate only when failure is imminent. An early warning capability, if existed, would greatly assist the effective management of structural fatigue in coordination with mission profile allocation and maintenance scheduling. However, the AE signals captured during the AE monitoring contain a wealth of information that is not properly exploited by the current AE practice which is solely based on counting hits. Some authors have taken a data-driven approach and applied statistical signal processing to extract standardized signal features but the success of these methods still depends on discarding the AEsignals not related to crack growth! We adopted a physics-based approach and directed our attention to understanding the origin andcausation of the wave signals recorded by the AE sensors and developing the computational models to assist this process. We found that AE spectrograms contain information that can be related directly to crack length. We posited that the AE waves traveling along the crack surface reflect at the other tip and create a standing wave pattern between the two crack tips. This standing wave pattern contains characteristic frequencies that are seen in the AE signal spectrum and may be directly related to crack length. Goal and objectives The ultimate goal of this project is to develop a baseline-free AE SHM methodology that does not require prior information about crack growth of load history. We aim to identify a direct correlation between crack length and AE signal and develop a method for extracting the crack length information from the AE signal features. The objectives of this proposed project are to (a) verify and experimentally validate that AE waveforms modify with crack length as predicted by preliminary FEM simulation; (b) develop a AI-enabled methodology for extracting the crack length information from the AE waveforms. Technical approachTo achieve the above goals and objective we propose a three-pronged technical approach: (1) Gather extensive experimental evidence (LCF-HCF) to verify/update the hypotheses (2) Use advanced modeling to generate AI training datasets of synthetic signals for known cracks plusnoise to (a) separate crack growth AE from other crack AE; and (b) estimate crack length from spectrogram information (3) Validated the trained AI on carefully conducted LCF and HCF experimentsAnticipated outcome of the researchThe anticipated outcome of the research would be an methodology for estimating the length offatigue crack from the AI-enabled analysis of the AE signalsemi0014-20-S-B001,specifically long-range fundamental scientific and engineering research for Non-DestructiveEvaluation, Structural Health Monitoring, Prognosis to address and mitigate fatigue and fracturedamage of Naval structural materials.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 06, 2021
- Source ID
- N000142112212
Entities
People
- Victor Giurgiutiu
Organizations
- Office of Naval Research
- United States Navy
- University of South Carolina