{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/124649"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/124649","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Nondestructive evaluation of material properties using physics-informed neural networks and mechanical waves","abstract":"Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-05-01","abstract_html":"Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-05-01","abstract_has_math":false,"creators":["Lee, Sangmin"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Civil Engineering","degree_department":null,"school":null,"contributors":["Popovics, John","Tartakovsky, Alexandre","Henschen, Jacob","Aggelis, Dimitrios"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-05","date_published":"2024-05","updated_at":"2026-07-22T22:25:02Z","subjects":["Concrete","Machine Learning","Material Characterization","Pinn","Nondestructive Testing","Ultrasound","Wave Propagation"],"languages":["en","eng"],"rights":["Copyright 2024 Sangmin Lee"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/124649","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Popovics, John","Tartakovsky, Alexandre","Henschen, Jacob","Aggelis, Dimitrios"]},{"key":"dc:creator","label":"Author","values":["Lee, Sangmin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-05","2024-04-25"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Civil Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Concrete","Machine Learning","Material Characterization","Pinn","Nondestructive Testing","Ultrasound","Wave Propagation"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2024 Sangmin Lee"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/124649"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-05-01","The student, Sangmin Lee, accepted the attached license on 2024-04-24 at 12:36.","The student, Sangmin Lee, submitted this Dissertation for approval on 2024-04-24 at 12:37.","This Dissertation was approved for publication on 2024-04-25 at 09:41.","DSpace SAF Submission Ingestion Package generated from Vireo submission #20323 on 2024-09-16 at 00:49:10","Characterizing in-place material properties is essential for quality control and condition assessment in construction and manufacturing. Recent advances in machine learning tools have increased the popularity of data-driven models to predict complex phenomena. Artificial neural networks (ANNs), which are popular machine learning tools, have achieved impressive results for image classification, time series prediction, and natural language processing tasks. However, data-driven models have limitations, such as the large size and complexity of ANNs needed to capture realistic behaviors, and the substantial amount of training data required, which significantly increases computational cost. Physics-informed neural networks (PINNs) offer a promising alternative that combine the strength of physics-based models with the flexibility of data-driven approaches. This dissertation presents a method for characterizing material properties using PINNs and mechanical wave-based data. The effectiveness of the proposed PINN models is evaluated using mechanical wave data from a variety of specimens, each with unique defects, material properties, and 1-D (rod), 2-D (plate) and 3-D (large slab) geometries that contain simulated defects. Bar waves, Lamb waves and Rayleigh surface waves are considered. Material properties such as wave velocity, quality factor, Young’s modulus and shear modulus are then predicted using PINN models. Mechanical wave propagation data are collected using a contactless sensing method, and the PINN models predict space-dependent wave velocity or Young’s modulus from these data, and the defects are indicated as low material property regions. Notably, the PINN models predict micrometer-scale cracks. Based on the material characterization work, several challenges to real-world applications were identified and addressed. One such challenge is the lift-off distance variation between a receiver and a targeted sample with contactless sensing. The lift-off variation obscures changes in material wave velocity over space. A proposed method using a laser displacement sensor compensates for the variation and yields accurate wave velocity profile along the specimen. The performance of PINNs using sparse wavefield data was also considered. Since contactless sensing methods use tens or hundreds of time averaged signals to increase the low signal-to-noise ratio, it can significantly reduce the data collection time in a field when only sparse data is needed to predict material properties. To compensate for the reduced amount of training data in sparse data, additional residual points are used in the developed PINN model so that the compatibility check with the governing equation can be performed in the area where the actual measurement was not performed. Lastly, PINN architecture that can handle multiple datasets is proposed. Because of poor data quality and accessibility issues in the field, the area of interest might need to be measured multiple times. A proposed method is able to utilize multiple datasets at once to improve the performance of space-dependent wave velocity prediction. Throughout this study, the developed PINN models show superiority over traditional signal processing methods or purely data-driven methods because the models are able to predict multiple inhomogeneous properties such as wave velocity, Young’s modulus, etc. using a single measurement dataset on different types of structures."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Nondestructive evaluation of material properties using physics-informed neural networks and mechanical waves"]}]}],"canonical_facts":{"dc:contributor":["Popovics, John","Tartakovsky, Alexandre","Henschen, Jacob","Aggelis, Dimitrios"],"dc:creator":["Lee, Sangmin"],"dc:date":["2024-05","2024-04-25"],"dc:description":["Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-05-01","The student, Sangmin Lee, accepted the attached license on 2024-04-24 at 12:36.","The student, Sangmin Lee, submitted this Dissertation for approval on 2024-04-24 at 12:37.","This Dissertation was approved for publication on 2024-04-25 at 09:41.","DSpace SAF Submission Ingestion Package generated from Vireo submission #20323 on 2024-09-16 at 00:49:10","Characterizing in-place material properties is essential for quality control and condition assessment in construction and manufacturing. Recent advances in machine learning tools have increased the popularity of data-driven models to predict complex phenomena. Artificial neural networks (ANNs), which are popular machine learning tools, have achieved impressive results for image classification, time series prediction, and natural language processing tasks. However, data-driven models have limitations, such as the large size and complexity of ANNs needed to capture realistic behaviors, and the substantial amount of training data required, which significantly increases computational cost. Physics-informed neural networks (PINNs) offer a promising alternative that combine the strength of physics-based models with the flexibility of data-driven approaches. This dissertation presents a method for characterizing material properties using PINNs and mechanical wave-based data. The effectiveness of the proposed PINN models is evaluated using mechanical wave data from a variety of specimens, each with unique defects, material properties, and 1-D (rod), 2-D (plate) and 3-D (large slab) geometries that contain simulated defects. Bar waves, Lamb waves and Rayleigh surface waves are considered. Material properties such as wave velocity, quality factor, Young’s modulus and shear modulus are then predicted using PINN models. Mechanical wave propagation data are collected using a contactless sensing method, and the PINN models predict space-dependent wave velocity or Young’s modulus from these data, and the defects are indicated as low material property regions. Notably, the PINN models predict micrometer-scale cracks. Based on the material characterization work, several challenges to real-world applications were identified and addressed. One such challenge is the lift-off distance variation between a receiver and a targeted sample with contactless sensing. The lift-off variation obscures changes in material wave velocity over space. A proposed method using a laser displacement sensor compensates for the variation and yields accurate wave velocity profile along the specimen. The performance of PINNs using sparse wavefield data was also considered. Since contactless sensing methods use tens or hundreds of time averaged signals to increase the low signal-to-noise ratio, it can significantly reduce the data collection time in a field when only sparse data is needed to predict material properties. To compensate for the reduced amount of training data in sparse data, additional residual points are used in the developed PINN model so that the compatibility check with the governing equation can be performed in the area where the actual measurement was not performed. Lastly, PINN architecture that can handle multiple datasets is proposed. Because of poor data quality and accessibility issues in the field, the area of interest might need to be measured multiple times. A proposed method is able to utilize multiple datasets at once to improve the performance of space-dependent wave velocity prediction. Throughout this study, the developed PINN models show superiority over traditional signal processing methods or purely data-driven methods because the models are able to predict multiple inhomogeneous properties such as wave velocity, Young’s modulus, etc. using a single measurement dataset on different types of structures."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/124649"],"dc:language":["en","eng"],"dc:rights":["Copyright 2024 Sangmin Lee"],"dc:subject":["Concrete","Machine Learning","Material Characterization","Pinn","Nondestructive Testing","Ultrasound","Wave Propagation"],"dc:title":["Nondestructive evaluation of material properties using physics-informed neural networks and mechanical waves"],"dc:type":["text"],"thesis:degree_discipline":["Civil Engineering"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:02Z"}