2020/12/20 by Matthew R. Jones, Timothy J. Rogers, Jones, Matthew R. +5 · 1 citation
Engineering · #Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Geophysical Methods and Applications #Machine Learning (cs.LG) #Sound (cs.SD) #Structural Health Monitoring Techniques #Ultrasonics and Acoustic Wave Propagation #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2012.11058
openalex publication_date 2020/12/20 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
In the field of structural health monitoring (SHM), the acquisition of\nacoustic emissions to localise damage sources has emerged as a popular\napproach. Despite recent advances, the task of locating damage within composite\nmaterials and structures that contain non-trivial geometrical features, still\nposes a significant challenge. Within this paper, a Bayesian source\nlocalisation strategy that is robust to these complexities is presented. Under\nthis new framework, a Gaussian process is first used to learn the relationship\nbetween source locations and the corresponding difference-in-time-of-arrival\nvalues for a number of sensor pairings. As an acoustic emission event with an\nunknown origin is observed, a mapping is then generated that quantifies the\nlikelihood of the emission location across the surface of the structure. The\nnew probabilistic mapping offers multiple benefits, leading to a localisation\nstrategy that is more informative than deterministic predictions or\nsingle-point estimates with an associated confidence bound. The performance of\nthe approach is investigated on a structure with numerous complex geometrical\nfeatures and demonstrates a favourable performance in comparison to other\nsimilar localisation methods.\n