Special Sessions
Guided Waves in Structures for SHM
Organisers: Wieslaw Ostachowicz and Annamaria Pau
Key words: sensors, sensing, SHM, damage detection, signal processing
Scope of Session: There is excellent potential for model-based approaches that utilize guided waves for damage detection, localization, and size estimation. This session covers key disciplines related to guided wave propagation in both isotropic and anisotropic materials. Authors are encouraged to submit papers that explore the phenomenon of elastic wave propagation, spanning a wide range of topics including linear and nonlinear behavior, 1D, 2D, and 3D propagation, time- and frequency-domain analyses, as well as experimental and numerical approaches in complementary investigations of structures. The proposed novel techniques may contribute to the efficient application of both local and global SHM technologies. The investigations outlined above aim to develop various strategies for diagnostics (damage size estimation and damage type recognition) and prognostics. The promising combination of these techniques should lead to innovative approaches that ensure safe operation.
Advances in Data Science and Artificial Intelligence for Structural Health Monitoring
Organizers: Mohammad Jahanshahi
Keywords: data science, data analysis, data fusion, machine learning, artificial intelligence, deep learning
Scope of Session: The recent advances in data science and artificial intelligence (AI) have led to ground-breaking innovations in the field of structural health monitoring (SHM). This special session aims to bring together researchers and practitioners to discuss the latest theoretical, computational, and experimental breakthroughs in applying AI techniques for structural identification, control, damage detection, and health monitoring. These innovative approaches can be applied to a variety of data types, including vibration signatures and non-destructive evaluation (NDE) data from advanced sensors. Topics relevant to this session include but are not limited to, machine learning-based damage assessment, deep learning, physics-informed machine learning and neural networks, generative adversarial networks (GANs), and other emerging data science and AI technologies with applications in SHM.
Advances in Computer Vision-based Structural Health Monitoring
Organizers: Mohammad Jahanshahi
Keywords: computer vision, deep learning, image and video processing, neural networks
Scope of Session: Advances in computer vision have unlocked new capabilities for structural health monitoring (SHM), driving the next revolution in information modeling and decision-making for the health management of structural systems. This session will provide the opportunity to discuss recent theoretical, computational, and experimental advances in applying computer vision techniques to structural identification, control, damage detection, damage localization, quantification, inspections, performance assessment, response measurement, and health monitoring. Topics relevant to this session include but are not limited to, deep learning, deep reinforcement learning, active learning, computer graphics simulations, innovative imaging for structures, image and video data collection and analysis, classification, convolutional neural networks, generative adversarial networks (GANs), transformers, quantification and localization, change recognition, displacement and dynamic measurements, sensor calibration, fusion and optimization, scene reconstruction, 3D LIDAR and depth sensors, robotics integration, and inspection and monitoring using UAVs and UGVs, along with other emerging computer vision and robotic technologies.
Nonlinear Ultrasonics and Topological Acoustics for Structural Health Monitoring
Organizer: Tribikram Kundu
Key words: nonlinear ultrasonics, topological acoustics, sensing, damage detection
Scope of Session: Papers are invited from various aspects of nonlinear ultrasonic and/or topological acoustics-based sensing techniques. Nonlinear ultrasonic techniques such as higher harmonic generation, sub-harmonic generation, frequency modulation, nonlinear impact resonant acoustic spectroscopy, sideband peak count - Index (SPC-I), vibro-acoustics and different wave modulation techniques as sensing techniques are becoming popular. How these techniques are used for non-destructive evaluation (NDE) and structural health monitoring (SHM) will be one research focus area of this special session. The second focus area will be topological acoustics based sensing techniques. Papers dealing with the advantages as well as shortcomings of various nonlinear ultrasonic and/or topological acoustics-based sensing techniques, and challenges encountered while implementing these techniques using body waves and/or guided waves are of interest for this session. Recent developments of new promising techniques that can overcome some of the existing shortcomings are of particular interest. Objective of this session is to give the attendees a broad overview of the current technology and recent developments of nonlinear ultrasonic and topological acoustics-based sensing techniques.
Acoustic Emission and Hybrid SHM
Organizers: Dr. Zhenhua Tian and Prof. Victor Giurgiutiu
Key words: acoustic emission, AE, non-destructive evaluation, NDE, structural health monitoring, SHM, passive detection, active detection, fracture, crack growth, composite, fiber breakage, matrix cracking, damage
Scope of Session: This special session will address the topic of acoustic emission and hybrid SHM. Acoustic emission (AE) is a passive SHM technique that relies on ‘listening’ to the elastic waves generated when an incremental crack growth occurs, or impact damage happens in composites. The elastic waves associated with AE events can travel a considerable distance in metallic structures which have a low damping dissipation coefficient. AE waves also travel in composite materials, but their travel distance may be less due to the higher damping dissipation of polymer matrix composites. Hybrid SHM techniques encompass a large class of methods that aim at combining several techniques to increase the probability of damage detection. For example, one may use passive SHM to record a damaging event (such as an impact in a composite structure) and then apply active SHM to try to estimate the magnitude of the resulting damage and its severity. Or one can listen to AE events which indicate that cracks are progressing into the structure and then follow up with the active SHM technique to evaluate the crack size. Or one can use two different active SHM techniques (e.g., pitch-catch wave propagation and electromechanical impedance standing waves) to better detect the damage location and size. But these are just examples. The session is open to all innovative techniques aimed at enhancing SHM capabilities. Contributions that judiciously combine theory and experiments are highly encouraged.
Seismic SHM for civil structures
Organizers: Maria Pina Limongelli and Mehmet Celebi
Key words: seismic SHM, civil structures, damage identification, real-time monitoring, emergency management
Scope of Session: During the last two decades, the need for seismic structural health monitoring (S2HM) by both property owners, as well as researchers and professionals, has evolved. As a result, numerous monitoring systems have been installed in structures in various seismic-prone countries that utilize real-time or near-real-time responses recorded during strong earthquakes to make informed decisions related to the health of structures. Data collected from S2HM systems have a strategic importance both for the advancement of knowledge on the behavior and performance of structures under strong seismic actions and for the calibration of realistic and reliable numerical models that are aimed to reproduce the structural behavior and to formulate a diagnosis about possible damages. Furthermore, the possibility of assessing the seismic vulnerability based on data recorded on the monitored structure opens new avenues in maintenance policies, shifting from a traditional ‘scheduled maintenance’ to a ‘condition-based maintenance’, carried out ‘on demand' or ‘automatically’, basing on the current structural condition. This Special Session aims to report recent advances in this field and successful applications for civil structures and infrastructures: buildings, bridges, historical structures, dams, wind turbines, and pipelines. The session deals with theoretical and computational issues and applications and welcomes contributions that cover, but are not limited to, seismic SHM algorithms for identification and damage detection, requisite strong motion arrays and real-time monitoring systems and projects, instrumentation and measurements methods and tools, optimal sensors location, experimental tests, integration of seismic SHM in procedures for risk assessment and emergency management. Such a session will provide a venue for the exchange of information on ongoing developments and assess the successes and limited successes of SHM.
Advancements in Smart Materials and Structures for SHM in Civil Engineering
Organizer: Yiska Goldfeld and Filippo Ubertini
Scope of Session: Advancements in smart materials and structural systems are revolutionizing structural health monitoring (SHM) in civil engineering, paving the way for a new era of intelligent infrastructure. Modern infrastructure increasingly demands durable, efficient, and cost-effective structural elements with multifunctional and self-sensory capabilities. These trends call for a new class of intelligent structures that utilize advanced processes and technologies to ensure real-time monitoring, damage detection, self-diagnosis, and predictive maintenance. This session aims to bring together researchers exploring the latest innovations in smart and multifunctional materials and structures, focusing on experimental and theoretical studies as well as practical applications. Topics include the development of smart sensors and actuators, self-monitoring structural elements, algorithmic strategies for self-sensory systems (including AI) and the integration of adaptive materials such as piezoelectric systems and self-healing composites in civil engineering structures. Through these advancements, participants will explore how novel SHM technologies are improving the safety, resilience, and sustainability of the built environment.
Machine Learning and Artificial Intelligence Enabled Health Monitoring of Composite Structures
Organizers: Xinlin P. Qing and Yongchao Yang
Key words: Sensor Network, Structural Health Monitoring, Machine Learning, Artificial Intelligence, Composite Structures.
Scope of Session: In the face of complex damage modes of composite structures and their service environments, most of current SHM methods have limitations to accurately and quantitatively monitor the damages in composite structures. With the rapid development of machine learning and artificial intelligence and their applications in SHM, it provides a great opportunity for more accurate and robust damage monitoring of composite structures in complex service environments. This session will provide a platform to discuss recent advances in leveraging machine learning and artificial intelligence techniques to enable structural identification, control, damage detection, inspection and health monitoring. Topics of interest include, but not limited to, machine learning, deep learning, active learning, transfer learning, physics-informed learning, meta learning, generative adversarial networks, and other new emerging machine learning and AI techniques for damage identification, assessment, and uncertainty quantification for composite structures.
Long-term damage identification in bridges under operational and environmental effects, including climate change
Organizers: Eloi Figueiredo, Ionut Moldovan, Luke J Prendergast, and Abdollah Malekjafarian
Key words: bridges, damage identification, machine learning, transfer learning, temperature, climate change
Scope of Session: The effects of operational and environmental variations (e.g., traffic loading, temperature, and flow characteristics of rivers) pose great challenges to structural health monitoring (SHM) of bridge assets from research to practice. In recent years, climate change has begun to pose increasing risk to the operational safety and health of bridges. Although the uncertainty associated with the magnitude of the change and the nature of how it might influence our built environment is large, the fact that our climate is changing is unequivocal. It is expected that climate change will manifest as another source of variability, resulting in changes in temperature, relative humidity, river flow, etc. Therefore, the main goal of this special session is to promote more coordinated and interdisciplinary research in the long-term vibration-based SHM of bridges affected by operational and climate variation, by proposing key developments in machine learning for damage identification under operational and environmental effects. Papers are welcomed on topics such as (but not restricted to):
- unsupervised and supervised machine learning for damage identification,
- transfer learning and domain adaptation,
- hybrid data sets from numerical models and/or monitoring systems for SHM,
- novel health monitoring algorithms,
- effects of climate change on the damage identification process.
Remote satellite-based structural health monitoring of structures and infrastructure
Organizers: Maria Pina Limongelli and Daniel Cusson
Key words: InSAR, space-borne, radar backscatters, deformation monitoring, damage detection
Scope of Session: Satellite-borne InSAR technology provides an appealing complementary approach to traditional SHM to measure mm-accurate displacements over large geographic areas, and to follow their evolution over time. The possibility to monitor large areas (e.g., an urban centre) opens new avenues for the developing automatic alerting systems that can flag several single structures with suspected structural integrity issues within a given network. For instance, InSAR measurements can provide information relevant to displacement time history, displacement rate, and thermal deformation, that can provide useful insight into ongoing deterioration phenomena. Optical-band satellite imagery can nicely complement InSAR-based monitoring with additional information for hydraulic structures, for example, on river current speed and direction, and nearby vortex formation, which can help assess the risk of pier scouring for river bridges and marine ports. This special session will provide the venue to present and discuss theoretical developments and field applications to foster future research collaborations on the topic. It welcomes contributions on algorithms (including AI-based) to process and analyze satellite imagery and data for damage detection purposes, case studies, measurement and calculation methods, and integration of remote and local sensing data for condition assessment and decision-making.
Recent Advances on Data Processing Techniques for Ultrasonics-based SHM/NDE
Organizer: Prof. Salvatore Salamone
Key words: guided waves, acoustic emissions, signal processing, SHM, NDE, damage detection
Scope of Session: This special session aims to collect and share recent developments in data processing techniques to enhance accuracy and capabilities of ultrasonic wave techniques for the SHM of complex structures. Authors are encouraged to submit papers topics that include but are not limited to: 1) deep learning, 2) data mining, 3) data analytics, 4) sparse matrices for machine learning. Both theoretical contributions and practical applications are welcome.
Wearable Sensors and Human Performance Assessment
Organizers: Ken Loh and Said Quqa
Key words: biomechanics, data visualization, digital health, human behavior, internet-of-things, physiological monitoring, prehabilitation, rehabilitation, sensing, textiles
Scope of Session: Individuals and teams are cornerstones of nearly every industry – hospitals, schools, corporate offices, schools, athletics, and the military. The successful, efficient, and effective execution of tasks – particularly physical-cognitive tasks to achieve a desired objective, such as hitting a golf ball, performing precision surgery, and putting out a building fire – require both individuals and/or the collective team to perform at their highest level. However, technologies that can measure and accurately quantify human performance are still lacking. The ones that do exist, such as commercial wearable sensors and Internet-of-Things (IoT) technologies, only provide gross information about human activity and may not be suitable for use in complex, field, or forward-deployed environments. Therefore, this special session is soliciting contributions focused on sensing the physiological and psychological conditions of human performance, as well as the interactions/interfaces between humans and the environment. Examples of specific topics of interest include: (1) wearable Internet-of-Things (IoT) technologies and feedback mechanisms; (2) bio-marker, biochemical, and bio-molecular sensing; (3) modelling of biological materials and systems; (4) static and dynamic characterization of human systems; (5) cognition and cognitive load measurements and modelling; (6) human digital twins; (7) human-structure interfaces that enhance system performance; (8) novel augmented/virtual reality and data visualization methods; (9) digital health systems and personal well-being technologies; (10) laboratory and field validation studies on human performance assessment; (11) artificial intelligence in human modelling, characterization, performance assessment, and performance enhancement; and (12) sports, military, medical, physical therapy, and biomechanical studies.
Integrating physics in data-driven methods for SHM
Organizers: Fotis Kopsaftopoulos & Dimitrios Zarouchas
Key words: physics-informed models, data-driven models, Explainable Artificial Intelligence (XAI), machine learning methods
Scope of Session: A successful implementation of SHM relies to a large extend on the quality of data and the damage information that this data contains. While the structure is in operation, multi-sensing techniques are usually employed (i.e., vibration-based, acousto-ultrasound, and optical-based techniques) in order to fulfill the four levels of SHM; damage existence, localization/identification, severity estimation (quantification), and remaining useful service life estimation/prediction, resulting to a vast amount of data that increases the complexity of analysis and, at the same time, reduces the efficiency and interpretability of the SHM system. Data-driven and Machine learning (ML) algorithms have attracted the interest of the community, and although promising results have been produced, they are still facing several significant challenges and their acceptance remains under consideration by the operators of structural assets and certification bodies. To overcome this challenge and provide competent solutions, integrating physics into the data-driven/ML SHM methods, in the form of prior expert knowledge, (semi)-empirical rules, physics laws and constraints, physics informed neural networks (PINN) offers a great potential. This special session welcomes contributions that explore and propose how integrating physics into data-driven methods/ML for SHM has the potential to increase the effectiveness, robustness, reliability, and deployment of SHM systems, as well as the interpretability of the health monitoring data and the explainability of data-driven/ML algorithms. Emphasis is placed on contributions that address the integration of physics- and data-driven methods within statistical and/or probabilistic frameworks as well as highlight comparative analyses using experimental data.