This PDF file contains the front matter associated with SPIE Proceedings Volume 12486, including the Title Page, Copyright information, Table of Contents and Conference Committee lists.

Nature through careful observation and tests of gliding avian species have resulted in new thoughts on how to design morphing uninhabited air vehicles (UAV) and what morphing motions might make for better performance. An understanding of avian flight stability suggests a new approach to morphing aircraft design. Of interest is how to create these motions using smart materials to replicate avian abilities. Coupled with new learning algorithms, methods for designing smart autonomous morphing airfoils for use in small UAVs are presented. Hardware based reinforcement learning (RL) techniques are used to teach a smart morphing wing to respond to gusts, following the inspiration of gliding gulls who respond immediately and autonomously to unknown changes in flow to maintain stability and control in unpredictable environments. We strive to translate this knowledge to flight control of UAVs. Last, a way forward is suggested to create new class of structures: autonomous multifunctional structures. An outline of what is needed in terms of future research is presented.

In recent years, there has been considerable interest globally in “smart cities,” which aim to improve the performance of urban systems and the experiences of citizens. However, growing interest in smart cities has given rise to many underlying challenges. Society is at a critical juncture where the decisions made to integrate technologies into daily life can either help create an equitable future, or will heighten the inequitable distribution of resources, knowledge, and power in society and infringe on privacy. Nowhere is this more notable than in civil infrastructure systems (e.g., transportation, social infrastructure, the grid, buildings), which are the foundation of society, provide basic public services to communities, and play a critical role in the distribution and usage of energy, goods, and mobility resources. Underpinning the management of many of these civil infrastructure systems is the spatio-temporal tracking of humans and the measurement of human-infrastructure interaction. Already, we have witnessed countless engagements where camera-based sensing systems are designed and deployed to track humans in public spaces. While camera-based solutions often promise to anonymize data by processing video footage using automated data processing tools, many communities are resistant to trust camera-based monitoring due to infringements on privacy and overarching notions of “Big Brother” within the community. Consequently, there is a need to track humans spatially and temporally in a way that minimizes reliance on privacy-invasive sensing technologies (e.g., cameras). In this paper, we design and exploit a human-tracking sensing network that leverages low-power, privacy-preserving passive infrared (PIR) sensors, which merit energy efficiency (i.e., no reliance on fixed/wired energy sources) and full privacy protection. To fill the gaps of existing work, we leverage the analog capabilities of PIR sensors to extract and quantify greater information–—direction of travel, distance relative to the sensor, and speed–—with a single PIR sensor, which would otherwise require multiple PIR sensors operating in a binary fashion in series to gain only partial information. Improved understanding and quantification of the information that can be extracted from the PIR sensor’s signal will pave the way for the development of human tracking networks that can explicitly consider tradeoffs between the value of information gained from a sensor and community-driven privacy concerns. To make this work more accessible by diverse research communities, the electrical software and hardware developed are open-sourced and detailed.

Hand gesture monitoring has aroused more and more interest with the development of emerging virtual reality technologies. High precision and resolution are needed for more accurate gesture simulation. Gesture acquisition can be obtained by different sensing technologies including elastomers, mark-tracking technology, and fiber-optic-based sensors. Among these mechanisms, fiber-optic sensors have their unique advantages due to their small size and high accuracy. Fiber Bragg gratings (FBGs) based fiber-optic sensor was commonly used to monitor the bending of joints of the finger. However, FBG arrays can only measure specific points therefore the pattern design and the choice of the location around the joints could be an issue. In this paper, we reported a distributed fiber-optic sensing system. Optical Frequency Domain Reflectometry (OFDR) technology was used to realize the distributed strain monitor along the whole finger. A pattern of straight lines was evaluated on the index finger and the real-time strain change can be monitored. Through the real-time strain response, this system was able to provide accurate strain data according to different gestures of fingers.

Structural health monitoring is crucial for ensuring the safety of civil infrastructure, and crack detection is an essential component of this process. Cameras provide high-resolution images of the structure’s surface, which can be analyzed to detect and locate cracks. LiDAR sensors use laser beams to scan the surface of the structure and produce detailed 3D point clouds that can be used to detect cracks and measure their dimensions. The proposed approach aims to improve the accuracy and efficiency of crack detection in SHM by integrating the complementary strengths of cameras and LiDARs in a simulation environment. The approach involves the use of an intelligent algorithm that can automatically fuse the data from the cameras and LiDARs to produce a more comprehensive and accurate representation of surface cracks. The algorithm uses a machine learning-based crack detection technique that can accurately identify and locate cracks in real-time. Furthermore, a depth camera is used to provide a denser point cloud than LiDAR of the crack. The integration of cameras and LiDARs for crack detection in SHM offers several advantages, such as improved accuracy, faster data acquisition, and reduced costs compared to traditional methods. The proposed approach addresses the challenges of data fusion, image processing, and intelligent algorithm development by offering a novel solution that leverages the strengths of both cameras and LiDARs. The findings of this study suggest that the proposed approach can significantly enhance the capabilities of SHM for crack detection. The approach offers a more accurate and efficient way of detecting cracks in real-time, which can help prevent further damage and ensure the safety of civil infrastructure.

In this paper, we proposed an IMU-based locomotion mode identification (LMI) system for ankle-foot prostheses. Specifically, an IMU sensor was mounted on the heel to collect the foot's dynamic information during walking. Then processing the dynamic data can estimate the foot trajectory for calculating the inclination grade of the terrain. It is noteworthy that our environment is constructed according to the inclination grade for ergonomics. For example, when the inclination angle ranges from 3 degrees to 11 degrees, the environment should be a ramp. On the other hand, when walking on different terrains, people prefer to move their feet around the ground's exterior. It is helpful for people to get the required foot clearance and, at the same time, minimize the energy needed for transporting the lower limbs. Therefore, with the estimated inclination grade, the presented method can precisely predict/identify the locomotion mode. Experimental results show that the average accuracy can reach 98.4% in five daily locomotion modes, including level-ground walking, stair ascending/descending, and upslope/downslope walking.

s study provides the development of a nondestructive yield strength estimation technique for metal 3D printed Ti-6Al-4V components using eddy-current measurement. Based on the relationship between the electrical conductivity and the grain size of the material, and the Hall-Petch relationship, the yield strength of metallic materials can be correlated by eddy-current phase value. First, a theoretical expression for the yield strength using the eddy-current phase value is explicitly derived. Then, for the experimental validation, the specimens with various yield strengths were produced by adjusting the cooling rate during the printing of each specimen. Then, the coefficients in the theoretical expression were estimated using the actual yield strength of specimens obtained via the conventional destructive tensile tests and the eddy-current phase values. Finally, the yield strength estimation performance was examined using the eddy-current signals obtained from test specimens with unknown yield strengths. The results indicate that the proposed technique can precisely estimate 3D printed Ti-6Al-4V components. The novelty of this study lies in (1) the derivation of an explicit relationship between the eddy-current phase value and the yield strength, (2) nondestructive yield strength estimation technique based on eddy-current testing, and (3) application to 3D printed Ti-6Al-4V plate specimens with various yield strengths.

There is an urgent need to better understand vehicle-rail interaction dynamics to pave the way for more consistent and automated rail crack detection methodologies, as opposed to relying on periodic and manual detection via track circuits or dedicated track geometry cars. Designing an open-source hardware framework for a lab-scale rail testbed would open the doors to further data collection and analysis needed to understand the dynamic response of cracked rails. We present a framework and the corresponding open-source hardware and software (published to GitHub) for developing a laboratory-scale motorized railroad testbed, with a vehicle that is modularly tuned to the dynamics of an in-service rail car.

This paper provides a comprehensive review on transducer technologies for underwater communications. The popularly used communication transducers, such as piezoelectric acoustic transducers, electromagnetic acoustic transducers, and acousto-optic devices are reviewed in detail. The reasons that common air communication technologies are invalid die to the differences between the media of air and water are addresses. Because of the abilities to overcome challenges the complexity of marine environments, piezoelectric acoustic transducers are playing the major underwater communication roles for science, surveillance, and Naval missions. The configuration and material properties of piezoelectric transducers effects on signal output power, beamwidth, amplitude, and other properties are discussed. The methods of code and decode communication information signals into acoustic waves are also presented. Finally, several newly developed piezoelectric transducers are recommended for future studies.

Sensory machine elements with integrated force sensors are paving the way for accurate load condition monitoring in machine tools. However, some machine elements face a fundamental measurement challenge: the intended high stiffness reduces any deformation during the operational lifespan drastically. In this study, an approach for high-resolution acquisition of elastic structure deformation under load in very stiff machine elements is presented. A novel optical image-based and structure-integrated deflection sensor allows a cost-effective two-dimensional deflection measurement with a measurement resolution of up to 90 nm and an accurate spatial resolution of the force direction. In particular, the change from local strain detection by strain gauge films to displacement measurements of reference planes allows for an increase of the measurement resolution. This paper discusses the relevant aspects and demonstrates the achieved measurement resolution by a biaxial sensory bending roll for the split profile bending process based on an optical image-based deflection sensor.

Sensitization refers to the formation of magnesium precipitates at the grain boundaries when Al-Mg alloy is exposed to elevated temperatures for extended periods of time, which can lead to reduced corrosion resistance and increased susceptibility to stress corrosion cracking. Since sensitization occurs at the microstructure level, a measurement technique that can measure multiple locations rapidly and is highly sensitive to minute changes is called for. This paper explores laser-generated longitudinal ultrasonic guided waves for sensitization detection in aluminum alloy plates. The samples were heat-treated to induce sensitization, and the degree of sensitization was determined by the nitric acid mass loss tests. The longitudinal ultrasonic guided waves were generated using a high-power pulsed free-space laser and detected using a continuous fiber laser. A digital signal processing algorithm has been developed to analyze the time-frequency components of the acquired signal. The experimental results demonstrated that the state of sensitization can be characterized by ultrasound parameters, such as attenuation.

This paper introduces laser excitation to identify the resonances of a bonded piezo-electric active wafer sensor (PWAS) that are associated with the longitudinal vibration of the host structure. The response of the PWAS to an impulse generated by a pulsed laser is converted into a time-frequency response using a digital signal processing algorithm. Based on the Lamb wave dispersion curves, the response of the PWAS to the symmetric mode (S0) and the antisymmetric mode (A0) can be separated in the time-domain. Identifying the structural mode associated with the resonances of the bonded PWAS simplifies the comparison between simulation and measurement and thus could potentially enable measuring the adhesive properties from the resonances of the bonded PWAS.

The authors aim to establish a technology for real-time measurement and correction of minute displacement of large optical structures in orbit. A test bed simulating a satellite-mounted telescope will be fabricated using a 70 cm ceramic mirror, and a demonstration test will be conducted applying the displacement compensation strut. First, as a preliminary verification of the demonstration test, an elemental test was conducted using a displacement compensation strut of the same size as the test bed. As a result, we confirmed that a control accuracy of ±10 nm could be achieved for a control range of 10 to 90 μm.

Fiber optic sensors are useful for Structural Health Monitoring (SHM) as they can sense environmental change and can react to external stimuli such as mechanical and thermal changes. Embedding Fiber Optic Sensors (FOS) in textiles provides some rigidity to the sensing mechanism as they can be fully integrated with the structures. Additionally, textiles with fiber sensors reduce the overall installation cost. Previously reported fiber optic sensors for traffic monitoring were not fully integrated with infrastructure and some sensors were discretely placed in the structure which prevents continuous data collection process along the entire fiber optic cable. In this study, distributed fiber optic sensor embedded in smart textile with a length of about 28m is presented and installed in a pedestrian bridge located at the University of Massachusetts Lowell with the objective of detecting vibration generated by pedestrians as they walk on the bridge. This paper demonstrates the load change variations in terms of corresponding strain change using Optical Frequency Domain Reflectometry (OFDR). The length of the test was approximately 2.5 hours, and strain changes were recorded at a 30-minute interval. During the test, for minimum traffic on the bridge at the testing time, the recorded strain value was around 16.2με. For larger loads, 2 people walking on top of the textile induced a larger strain change which the record value was 371.2με. Based on the load of the bridge, strain changes results depict that higher loads results in higher strain change and vice versa. This type of distributed fiber optic sensor can be used for the application of real-time traffic monitoring as well as to continuously monitor the structure status of the infrastructure.

This study investigates the heat treatment effect on the Rayleigh scattering based distributed single-mode fiber optic temperature measurement. The temperature increment and sustained time at the peak temperature considered in the test were 20 ℃/30 ℃ and 90 min/60 min, respectively. Moreover, a finite element model was established to investigate the effect of the thermal expansion of the coatings on the optical fiber core strains and thus temperature sensitivity before the coatings soft and melt. A theoretical derivation for the fiber core strains by using Lame equations was performed to verify the accuracy of the finite element model. It is found that the one-time heat treatment successfully eliminates the hysteresis effect and stabilizes the Rayleigh scattering based distributed temperature measurement up to 1000 ℃ along the optical fiber regardless of the temperature increments. The sustain time at the peak temperature does not significantly affect the Rayleigh scattering measurement. Moreover, a unified Rayleigh frequency-temperature equation (with R2 more than 0.999) is fitted for fiber optic temperature measurement after the heat treatment with different temperature increments. In addition, the numerical fiber core strain at ambient temperature provides the upper limit as compared to the experimental results at high temperature, and the thermal-induced strains on the fiber core are smaller than 1 με. Therefore, the typical dual-layer coating effect on the temperature sensitivity of the Rayleigh frequency shift can be neglected at high temperature for civil engineering applications. Finally, parametric studies are performed to investigate the effect and sensitivity of material properties and geometric parameters on the fiber core strain, based on the validated finite element model. The present study is further promoting Rayleigh scattering based fiber optic temperature measurement application.

Optical fibers are reliable sensors and can provide accurate measurements in microfluidic systems. Photonic crystal fibers are a variation of the conventional optical fiber, where the difference is the distribution of an array of hollow cores along the length of the fiber. These types of fibers are used as sensors in microfluidic cell cultures and offer the unique accessibility of using the hollow cores as a transport mechanism for biological samples. This type of system can benefit the biomedical and healthcare industry by reducing unnecessary manufacturing costs, while modeling complicated systems such as the interaction between the blood and cancer cells in the circulatory system. This manuscript briefly reviews recent literature on optical fiber sensors exploiting the hollow core technology of these optical fibers in real-time monitoring applications while transporting biological compounds. The significance of using biological samples such as human or animal cells integrated with optical fiber sensors can enhance research and clinical testing in modern industry. The remaining portion of the manuscript is an analysis on an experimental system designed to transport a cell culture through a photonic crystal fiber that is connecting two microfluidic devices. The internal diameter of the hollow core of the fiber is 22 µm, and the diameter of the cell is approximately 15 µm, indicating no extreme limitations in transport. Spectroscopic data confirms the presence of a cellular aggregate impeding the flow rate either within the hollow cores of the fiber, or the region transitioning between the microfluidic device and the fiber. The analysis consists of determining the mechanisms that contribute to cellular aggregation in microfluidic theory, to accurately model this phenomenon. The introduction of the analysis examines the experimental configuration with a class of dimensionless quantities that are often used in microfluidic theory. The Reynold’s, Peclet, capillary, Knudsen number, and Transcapillary conductance are dimensionless quantities used in recent literature in describing microfluidic transport. An analysis on these quantities led to the Lattice Boltzmann method for modeling the behavior in the experimental configuration. The Lattice Boltzmann method is a method in computational fluid dynamics that operates by imposing a discrete lattice to stimulate the fluid environment. The system evolves through collision and streaming processes, and along with Brownian dynamics, dictates the behavior of deformable particle mechanics. The combination of these two models is applied to the problem cellular aggregation in the experimental configuration. A simulation depicting the scenario of cellular aggregation in the hollow core fiber under conditions of Poiseuille flow with pressure differential of ∆p = 10² Pa, fluid velocity U = 10³ µm/sec, with a particle diameter to constriction diameter of 10/15 measured at the microscale (10⁻⁶ m). This is in comparison with conditions the experimental system operated at under a pressure differential ∆p = 1 bar, volumetric flow rate Q = 1.2 µL/min, at room temperature, with a particle to constriction diameter of 15/22. The analysis as approximated by the dimensionless quantities provides support that modeling aggregation in microfluidics is acceptable under the Lattice Boltzmann and Brownian dynamics method. Future directions of those research include repetition of these transports under varying conditions, to determine an environment for transport to occur. An examination into molecular interactions and properties, such as pressure differential, surface characteristics, and excluded volume, affecting aggregation and fluid flow is also to be considered.

This article presents a physics-based structural health monitoring (SHM) approach applied to a pretensioned adjacent concrete box beams bridge to predict the deformations associated with the presence of transient loads. A detailed finite element model was generated using a commercial software to create an accurate model of the bridge. The presence of concentrated loads on the deck at different locations was simulated, and a static analysis was performed to quantify the deformations induced by the loads. Such deformations were then compared to the strains recorded by an array of wireless strain gages during a controlled truckload test performed by an independent third-party. The test consisted of twenty low-speed crossings at controlled distances from the bridge parapets using a truck with a certified load. The array was part of a SHM system that consisted of 30 wireless strain gages.

Coastal flooding events have caused many issues to infrastructure including bridges and highways. How to assess the flooding level and infrastructure damages in a low-cost, rapid, and accurate approach is critical to the infrastructure performance recovery. Due to the limited access to infrastructure during the post-flooding events, it is very challenging to evaluate infrastructure conditions closely. With the help of small unmanned aerial vehicles and onboard cameras, it provides the possibility to inspect the infrastructure conditions from images captured by drones remotely. With the additional help of image processing algorithms, it can help capture the infrastructure conditions and flooding levels from the imageries automatically with post-processing analysis. In this paper, we apply several different image processing algorithms to assess the infrastructure conditions by segmenting the flooding zone from the infrastructure. The performance of these algorithms in assessing infrastructure conditions is compared based on different factors with previously taken airborne imageries of infrastructure and flooding events. The performance of image processing is summarized and future work of assessing the infrastructure post-flooding damages is discussed.

The long-term behavior of concrete is driven by rheological effects and impacts the safety and serviceability of high-rise buildings. Rheological effects are difficult to predict due to their dependence on environmental conditions and loading history. Recently, data-driven prediction using long-term structural health monitoring data have shown success in prestressed bridges. In this work, calibration of rheology models using long-term monitoring data towards forecasting of long-term behavior of high-rise buildings is investigated. A calibration strategy is identified that enables improved long-term forecast. The method is evaluated on data from two residential high-rise buildings.

Existing automated concrete inspection methods are intractable: capturing images under ambient conditions which can vary substantially. Furthermore, an opportunity may have been overlooked: utilizing illumination techniques to enhance defect contrast during imaging which may improve automatic defect detection accuracy. In this work, we present a robotic-mountable lighting apparatus that implements contrast enhancing illumination techniques in an automated package in order to improve crack detection and classification in concrete. Geometrical lighting techniques; directional and angled, were tested on three cracked concrete slab samples. Results from blind/reference less image spatial quality evaluation (BRISQUE) show that both directional and varied angled lighting influence the quality in different associated regions in an image. Furthermore, the region-based crack detection algorithm Faster R-CNN attained a higher accuracy when images were enhanced with directional lighting during all samples tested. The direction of highest accuracy was not consistent over samples, and is likely dependant on features such as crack location, width, orientation etc. This emphasises the importance of adaptive lighting: illuminating the surface with the most suitable conditions based on an initial observation of the feature or defect. This system represents the initial step in a fully automated and optimised concrete inspection system capable of defect capture, classification, localization and segmentation.

Fiber-reinforced plastic (FRP) is a material used to reinforce civil engineering structures. Carbon-fiber-reinforced plastic (CFRP) is composed of conductive reinforcing filler and dielectric matrix. By utilizing the internal structure of CFRP and the difference in electrical conductivity of the components, carbon fiber (CF) and resin matrix can serve as a structural electrode and a frictional material, respectively. Electrostatic charge is generated by friction between resin covering CF and external materials. It induces an alternating current through CF according to the distance change between charged layers. Then, CFRP can be applied as a structural sensor using the triboelectric effect. In this research, we identified the triboelectric effect occurring on the surface of the composite and the electrostatic induction phenomenon occurring inside the composite. In addition, the voltage signal changes according to the movement of external materials. By identifying the effects that occur in composite materials, we have confirmed the possibility of realizing a smart civil engineering structure that can detect touch by itself.

The actual tensile force of pre-stressed (PS) tendons of a pre-stressed concrete (PSC) girder is one of the important factors for evaluating the performance of PSC girder bridges. However, it is hard to measure the residual tensile force of outdated PSC bridges because the PS tendons were buddied in the concrete and the any sensory systems were not installed. To measure the tensile force of the outdated PSC girder bridges, this study proposed an external magnetization based tensile force estimation method using external magnetization sensor(EMS). The magnetic hysteresis of PS tendons is changed according to the residual tensile force. To measure the magnetic hysteresis of outdated PSC girder, the EMS was designed to concentrate the magnetic field to the PS tendons which located inside of girder. The 8 magnetization coils were installed to the U shape frame which can cover the both side of PSC I-shape girder and the designed currents were inputted to each magnetization coils to concentrate magnetic field. The flux density of magnetized PS tendon was measured hall sensors which located at the frame. To verify the proposed method, the in-field tests were performed. The magnetic hysteresis of PSC girder with difference tensile forces were measured using EMS and the feature was extracted to estimate residual tensile force of PSC girder. The estimated residual tensile forces were compared with reference tensile forces measured by hydraulic jacking machine which installed anchorage of PSC girder specimen. According to the measurement results, the proposed method can be a one of the solutions to measure the residual tensile force of outdated PSC I-shape girder bridges.

Currently, the glass curtain wall has become very popular in contemporary architectural wall decoration method because of its aesthetics, lightweight, energy saving, and thermal insulation. However, the damage of the glass curtain wall is inevitable due to its material nature. Currently, the detection method of glass curtain wall is to use regular manual detection, It highly depend on the experiences of the inspector and are not real-time monitoring method. Therefore, it is necessary to develop a monitoring method for the evaluation of status of glass curtain wall, which can realize the real-time monitoring and high reliability. This paper proposed a combined acoustic emission and vibrational modal analysis method to achieve multi-scale damage detection for glass curtain wall: Modal analysis is used to detect structural silicone sealant failure, bolt loosening and corrosion of glass curtain wall, which refers to the first-step inspection to approximately determine the damage status. And Acoustic emission (AE) is further used to continuously monitor the glass curtain wall to provide more detailed damage evaluation. The proposed scheme is verified by COMSOL Multiphysics 5.5. The relationship between the damage degree of structural silicone sealant, bolt failure and the modal frequency of the glass panel was also obtained. And also, AE has been proved to be able to realize real-time monitoring and early warning of objects. Therefore, the feasibility of the proposed design has been fully explained. It provides an option for glass curtain wall inspection.

Concrete is the most widely used material in civil engineering, which has many advantages in terms of strength, accessibility, and durability. Currently, the high-strength concrete has been used in some projects. However, its failure mode is more complex than the regular concrete. Especially for the loading rate, it may significantly affect the failure mode. Therefore, it is essential to quantify the process of the failure for better application of high-strength concrete with consideration loading rate. Due to many merits of acoustic emission (AE), it has been successfully used for evaluating the progress of concrete damage. Therefore, in this study, the experiments were proposed to correlate the AE with failure mode in the high-strength concrete. As progress of the damage, the AE activities show highly related to the mechanical response. Additionally, the loading rate directly related to the failure mode is also considered. Compared with the regular concrete, the failure of the high-strength concrete shows more brittle behavior. And the phase of AE signals has been changed from obvious three stages to two stages with increase of loading rate, which can be used for further identification of failure mode.

Deterioration modeling is a critical task in bridge maintenance planning. Advanced deterioration modeling allows maintenance actions to focus on where and when they are most needed. In this paper, we use a neural network survival model applied to National Bridge Inventory (NBI). Survival modeling is a technique of modeling the time-to-event and has been traditionally associated with fields such as medical sciences and reliability engineering. In this work, we focus on understanding the effect of bridge population heterogeneity on model performance. Multiple structural systems, materials, and deck protection methods exist within the bridge population. In addition, bridges are subjected to various environmental and loading conditions. We expect that heterogeneity of the population will provide difficulties in selecting a suitable model. To understand this problem, we study the effect of heterogeneity in the bridge population on model prediction performance. To do this, we first split the dataset into subsets. We approach the problem of dividing the data from two unique angles: statistical clustering methods and a physics-based approach where we split the dataset into subsets based on the underlying deterioration mechanisms. After splitting the data into subsets, we fit separate survival models on each of these subsets and compare the prediction performance with the other subset models and a model fitted on the entire dataset. This comparison allows us to understand if the type of survival models we have utilized is more suitable for some deterioration mechanisms and structural types than others.

This research analyzes the sensitivity of the ultrasound features to sensitization, a microstructural damage developed in Al-Mg alloys when subjected to extended period of heat treatment. For rule extraction, an Adaptive Neuro-Fuzzy Inference System (ANFIS) was first developed and customized to house the ultrasound features from the Electromechanical Impedance (EMI) of Piezoelectric Wafer Active Sensor (PWAS) attached on aluminum plates with progressive sensitization development. The features are rooted in the rise time and the amplitude of the time domain signal that was obtained at two different resonance phases. The extracted fuzzy rules were analyzed to determine the sensitivity of each feature at both the model and rule level. The analysis revealed that the rise time changes at higher resonance frequency and the amplitude change at lower resonance frequency has a stronger correlation to sensitization. The results of rule significance evaluation demonstrated that the data measured at low resonance frequency exhibits more correlation to sensitization.

The potential of levee failures poses significant risks to populations living behind them. Levee monitoring using ground velocity measurements obtained from geophones has been demonstrated with the simultaneous deployment of wired geophone arrays. However, the scale of levees makes their monitoring with wired sensors a challenging task. This work reports on the development of a stand-alone geophone monitoring system for levees constructed of earthen embankments. The newly developed open-source sensor package can simultaneously measure ground velocity, conductivity, and temperature in addition to ambient atmospheric pressure and humidity. The system is fully independent of processing, power management, sensors, and data storage all contained within a single instrument. This work reports the initial experimental validation of the proposed system using a granular earthen levee in a flume under controlled erosion conditions. Data is collected and post-processed for anomaly detection; sensing capabilities, and the effect of sensor noise are discussed. To the knowledge of the authors, this is the first open-source stand-alone geophone system developed and tested for the monitoring of earthen levees.

This paper studies time-delayed simultaneous input and state estimation to enhance estimation accuracy for systems without direct feedthrough, such as earthquake-excited building structures, using absolute floor acceleration measurements. Rank matching, strong observability, and invertibility conditions are crucial for the stability and convergence of input and state estimation. Real-time approaches have achieved successful estimations when those conditions are satisfied. However, a dynamic system model often does not hold those conditions when using acceleration measurements, leading to significant errors in the estimations. To this end, the authors recently developed an optimal sensor placement algorithm to ensure the system model holds the above conditions to achieve accurate real-time estimation. However, accurate estimation in some cases remains challenging because of incomplete measurements, modeling error, and measurement noise. This paper proposes an extended time-delayed joint input and state estimation algorithm (ETDIS) based on the invertibility matrix. Specifically, by incorporating the prior knowledge of the input, the proposed ETDIS is designed from a Bayesian perspective, considering measurement noise to enhance estimation accuracy. In particular, the innovation is used to obtain the input estimation, which is interconnected with the state space equation for state estimation. ETDIS relaxes the rank-matching condition and is more robust against the lack of conditions. Accurate online input and state estimation with a delay and satisfactory computational cost is achieved by utilizing the proposed ETDIS and limited acceleration measurements. Numerical studies are presented to verify the effectiveness of the proposed method.

This paper presents the implementation of a long-term wireless ultrasonic thickness measurement system on an in-service steel bridge. Developed on the Martlet wireless sensing platform, a wireless ultrasonic thickness measurement device is added with the capacity of high-voltage excitation, filtering/amplification of the received ultrasonic signal, and high-speed sampling (~80 MHz). Martlet ultrasonic thickness devices as well as 2.25MHz dual-element transducers are installed on a steel bridge in western Georgia, USA. The thickness measurements are taken at scheduled time intervals for long-term monitoring. A gateway server installed on-site collects ultrasonic data samples from Martlet units and uploads data into the cloud. Thickness values are calculated automatically using the autocorrelation of the ultrasonic data.

Icing is a well known phenomenon in aviation. It happens, when an aircraft is flying in a freezing air mass containing water droplets which are then accreting to the surface of the aircraft. If the ice accretion gets too severe, it reduces the aerodynamic performance and increases the weight which may lead to a catastrophic failure of the aircraft. Recognizing an ice accretion is essential to initiate countermeasures against icing e.g. switching on ice protection systems or leaving the icing zone. By now, this mostly relies on the pilot looking to the front fuselage to detect ice accretion, which may be too late. To overcome this, ice accretion sensors can be used. Two types of ice accretion sensors have therefore been developed and tested at the German Aerospace Center (DLR). The first type is based on sending ultrasonic lamb waves through an icing prone structure, e.g. the leading edge of an airfoil. If an ice accretion happens, this will alter the waveguide characteristics of the structure. This change in the waveguide parameters hence influences the sensor signal, which can be detected by the sensor electronics. For the lamb wave sensor, piezoelectric transducers are therefore applied on the structure at a certain distance since the lamb wave signal needs a minimum amount of traveling time. The second sensor type also uses piezoelectric transducers and is based on measuring the impedance of the transducer. Due to the electromechanical coupling via the piezoelectric effect, the measured impedance of the transducer also contains the characteristics of the structure where it is applied to. Since an ice accretion will change these structural characteristics, measuring the impedance generates an obtainable sensor signal. Both of these sensor principles have been tested in a icing wind tunnels on a typical airfoil structures. There, the impedance sensor was mounted on three different positions in chord direction. The lamb wave based sensor has been mounted on the leading edge and further downstream positions to detect an ice accretion along the flow. During the icing tests multiple icing runs have been performed. It could be shown, that both sensors are capable of detecting the beginning of the ice accretion and the presence of ice on the structure. The lamb wave sensor reacts nearly instantaneously to the presence of ice. According to preliminary data, the impedance sensor needs a slightly larger ice thickness to produce a signal, but can handle larger ice thicknesses compared to the lamb wave sensor.

The battery-powered wireless sensor network (WSN) is a promising solution for structural health monitoring (SHM) applications because of its low cost and easy installation capability. However, the long-term WSN operation suffers from various concerns related to uneven battery degradation of wireless sensors, associated battery management, and replacement requirement, and ensuring desired quality of service (QoS) of the WSN in practice. The battery life is one of the biggest limiting factors for long-term WSN operation. Considering the costly maintenance trips for battery replacement, a lack of effective battery degradation management at the system level can lead to a failure in WSN operation. Moreover, the QoS needs to be ensured under various practical uncertainties. Optimal selection with a maximal number of nodes in WSN under uncertainties is a critical task to ensure the desired QoS. This study proposes a reinforcement learning (RL) based framework for active control of the battery degradation at the WSN system level with the aim of the battery group replacement while extending the service life and ensuring the QoS of WSN. A comprehensive simulation environment was developed in a real-life WSN setup, i.e. WSN for a cable-stayed bridge SHM, considering various practical uncertainties. The RL agent was trained under a developed RL environment to learn optimal nodes and duty cycles, meanwhile managing battery health at the network level. In this study, a mode shape-based quality index is proposed for the demonstration. The training and test results showed the prominence of the proposed framework in achieving effective battery health management of the WSN for SHM.

Fault parameters in a structure are identified by matching measurements with model predictions in the parametric space. As high frequency measurements are preferred to uncover small-sized damage, piezoelectric impedance/admittance active interrogation has shown promising aspects. Nevertheless, challenges remain. The amount of useful measurement information is generally insufficient to pinpoint damage. The inverse identification is usually underdetermined. In this research, we develop a combinatorial enhancement to tackle these challenges. A tunable piezoelectric impedance sensing procedure is developed in which an adaptive inductor element is integrated with the piezoelectric transducer, which will lead to significantly enriched measurement data for the same damage. Subsequently, an intelligent learning automata-based multi-objective particle swarm optimization framework is synthesized to inversely identify the damage location and severity. Case studies are conducted to highlight the accuracy of the damage identification.

Electrical impedance tomography (EIT) is a non-invasive method of spatially mapping the electrical conductivity distribution of a domain based on a limited number of externally collected voltage-current measurements. This modality has been widely explored in the state of the art for damage detection, shaping, and localization in conductive composites (e.g. continuous carbon fiber composites and various nanofiller-modified continuous glass fiber composites) for purposes such as nondestructive evaluation (NDE), structural health monitoring (SHM), and embedded sensing. Mathematically, EIT is an ill-posed inverse problem that requires regularization to solve. To date, materials-focused practitioners of EIT have used relatively simple forms of regularization including, among others, Tikhonov regularization and the discrete Laplace operator (i.e. a smoothness prior). This is limiting because much more advanced types of regularization exist and have potential to significantly improve EIT for material state awareness. Therefore, in this work we propose and experimentally validate a novel mixed form regularization for the EIT inverse problem. In this approach, the discrete Laplace operator or smoothness prior is combined with a conditionally Gaussian prior (i.e. a focal prior). This mixed formulation has the benefit of simultaneously filtering out oscillatory background conductivity perturbations (via the smoothness prior) while still permitting outliers in the solution space (via the focal prior), which is expected to be the case for highly localized damage features in a background of otherwise zero change. The proposed mixed formulation was experimentally validated on two different three-dimensional composites: a carbon black (CB)-modified glass fiber/epoxy tube and a carbon fiber/epoxy laminate shaped as a representative NACA airfoil. Both composite specimens were subject to low-velocity impact damage via a drop-tower rig. It was found that the mixed smoothness + conditionally Gaussian regularization approach markedly outperforms the traditional smoothness only regularization, which allows for much clearer visualization of the damaged state of the material. This work demonstrates the importance of researching advanced regularization methods for materials imaging via EIT.

Structural Health Monitoring (SHM) can provide valuable information for maintenance-related activities and post-disaster emergency management. However, as with any technological system, SHM systems can be susceptible to errors due to malfunctioning. Therefore, it is essential to assess the potential for malfunctions and the associated costs of maintenance and repair when evaluating the long-term benefits of SHM systems. In the last two decades, sensor validation tools (SVTs) have been proposed to support decisions by isolating and discarding abnormal data. Recently, the authors of this paper have proposed a framework based on the Value of Information (VoI) from Bayesian decision analysis to account for different states of an SHM system and assess the benefit of SVT information. By quantifying the additional value obtained from SVTs, decision-makers can make more informed decisions about investing in these systems. This framework is here demonstrated on a real case study, namely the S101 bridge in Austria, which has been artificially damaged for research purposes. The benefit of collecting SHM and SVT information is quantified by considering a simple decision problem related to the management of the bridge in the aftermath of a damaging event. Overall, the study highlights the potential benefits of using SVTs to improve the reliability of SHM data and inform decision-making in the management of structures.

This study uses a novel directional lighting approach to produce a computationally efficient five-channel Visual Geometry Group-16 (VGG-16) convolutional neural network (CNN) model for concrete crack detection and classification in low-light environments. The first convolutional layer of the proposed model copies the weights for the first three channels from the pre-trained model. In contrast, the additional two channels are set to the average of the existing weights along the channel. The model employs transfer learning and fine-tuning approaches to enhance accuracy and efficiency. It utilizes variations in patterned lighting to produce five channels. Each channel represents a grayscale version of the images captured using directed lighting in the right, below, left, above, and diffused directions, respectively. The model is evaluated on concrete crack samples with crack widths ranging from 0.07 mm to 0.3 mm. The modified five-channel VGG-16 model outperformed the traditional three-channel model, showing improvements ranging from 6.5 to 11.7 percent in true positive rate, false positive rate, precision, F1 score, accuracy, and Matthew’s correlation coefficient. These performance improvements are achieved with no significant change in evaluation time. This study provides useful information for constructing custom CNN models for civil engineering problems. Furthermore, it introduces a novel technique to identify cracks in concrete buildings using directed illumination in low-light conditions.

This study developed a framework for real-time dynamic analysis of structural members using physics-informed neural networks (PINN). The interest in the use of augmented reality (AR) and virtual reality (VR) technologies to visualize the results of simulations is now increasing, and many researches are taking efforts to make these simulations more interactive and real-time. However, the application of structural dynamic simulations is limited due to its high computational cost. In this study, the Physics-informed neural networks (PINN) was used to conduct the real-time vibration analysis of a cantilever beam as a basic investigation. Prior to the real-time simulation, a PINN model for solving the cantilever beam undamped free vibration problem was constructed. Sequential trainings and predictions for the real-time simulation were then implemented at fine increment time steps by PINN. The distributions of displacement and bending moment, which were the outputs of the PINN simulations were visualized in AR on the real beam with converting the outputs to color contour for intuitive understanding. The RS framework based on PINN simulation and AR was then recognized to lead to the RS with data assimilation for real-time evaluation of structural condition using measurement data.

The increasing number of bridge collapses has progressively fueled the interest towards the development of monitoring strategies able to ensure real-time damage assessment and preserve structural integrity with correct maintenance. In the context of vibration-based Structural Health Monitoring, recent improvements in sensor technologies and computer science have encouraged the use of Machine Learning algorithms in many engineering fields. In this light, the described methodology proposes to combine Domain Adaptation with supervised learning methods, such as the K-Nearest Neighbors algorithm, in order to correctly assign damage labels to statistically-aligned features. To this aim, natural frequencies gathered during healthy and abnormal conditions are selected as damage-sensitive quantities, bringing a physical meaning and providing information on global structural dynamics. Damage detection results are evaluated and compared before and after Domain Adaptation by employing specific performance metrics. The developed procedure is validated on the Z24 benchmark bridge and the Finite Element Model of the same structure, properly calibrated based on available experimental data. Within the numerical environment, several modal analysis are carried out both assuming pristine conditions and simulating realistic damage scenarios, that involve concrete elastic modulus’ reduction for specific bridge elements. The goal of the proposed technique would be to effectively identify and classify different types of damage cases, enabling knowledge transfer among a population of structures and thus representing a prompt engineering decision-support tool to capture damage-induced variations.

Surface strain sensors, such as linear variable differential transformers, fiber Bragg gratings, and resistive strain gauges, have seen significant use for monitoring concrete infrastructure. However, spatial monitoring of concrete structures using these sensor systems is limited by challenges in the surface coverage provided by a specific sensor or issues related to mounting and maintaining numerous mechanical sensors on the structure. A potential solution to this challenge is the deployment of large-area electronics in the form of a sensing skin to provide complete coverage of a monitored area while being simple to apply and maintain. Along this line of effort, networks constituted of soft elastomeric capacitors have been deployed to monitor strain on steel and composite structures. However, using soft elastomeric capacitors on concrete surfaces has been challenging due to the electrical coupling between the sensors and concrete, which amplifies transduced strain signals obtained from the soft elastomeric capacitors. In this work, the authors investigate the isolation of the soft elastomeric capacitors from the concrete by extending the styrene-block-ethylene-co-butylene-block-styrene matrix of the soft elastomeric capacitors to include a decoupling layer between the electrode and the concrete. Experimental investigations are carried out on concrete specimens for which the soft elastomeric capacitor is adhered to with a thin layer of off-the-shelf epoxy and then loaded on the dynamic testing system to monitor strain provoked on the concrete samples. The results presented here demonstrate the viability of the electrically isolated soft elastomeric capacitors for monitoring strain on concrete structures. Initial comparisons between un-isolated and electrically isolated soft elastomeric capacitors showed that the nominal capacitance of the soft elastomeric capacitor is significantly lowered by adding an isolation layer of SEBS. Furthermore, strain results for the soft elastomeric capacitors are compared to ones from a resistive strain gauge and digital image correlation. The data obtained is significant for modifying soft elastomeric capacitors with the anticipation for future use on concrete structures.

Recent wildfire events in California and Oregon have resulted in localized water contamination. A potential cause is the heating of polymer-based water service lines and mains in these communities. Identifying the source of contamination can be a huge burden on municipalities, taking significant resources and time. The investigation in Santa Rosa and Paradise, California took months and millions of dollars, and delayed their recovery. These contamination events highlighted the need for a quicker, more efficient way to check water lines in affected areas. Previous research has shown that the threshold temperatures that result in contamination are 194° C for polyvinyl chloride (PVC) and 250° C for high-density polyethylene (HDPE) pipes, respectively. The objective of this work is the development of a low-cost sensor system to identify potentially damaged pipelines and sources of water contamination. The proposed solution is a radio frequency identification (RFID) based temperature sensor to indicate once a certain temperature is reached. Passive, ultra-high frequency RFID tags are used in conjunction with a trigger mechanism that disconnects after the threshold temperature is reached over a meaningful duration. Passive RFID will allow for the tags to operate without the use of batteries, and are very affordable. The design and characterization of the sensor utilizes two experimental frameworks: (1) benchtop testing, (2) small-scale tests in a more realistic environment. The benchtop testing identifies the trigger temperature, mechanism, and reliability of the sensor design. The small-scale testing installs the sensors on buried pipes subjected to a realistic fire load. The resulting design and characterization will be presented in terms of accuracy, durability, and reliability. Additionally, the heat flux of the benchtop testing will differ from a more realistic environment, so the results will be compared to isolate how the heat flux might impact future wildland-urban interface (WUI) based sensor development.

This paper aims to investigate the performance of piezoelectric sensors with different shapes of 3D-printed microstructures. Based on the numerical analysis in the time-frequency domain, the microstructures are printed directly on the PVDF transparent film exhibiting higher piezoelectric coefficients using a high-resolution two-photon polymerization method. Bi-directional gold IDTs are fabricated by sputtering gold onto the substrate surface using a 3D-printed stencil. The mechanical properties of the film and surface morphology of printed microstructures are examined using a nanoindenter and a 3D profilometer. The change in frequency response due to the microstructure is measured using a network analyzer. This study will be a reference for developing an efficient wave-based gas sensor with enhanced sensitivity.

This project aim to robotically deploy vibrating wireless strain gauges (VWSG) into small scale steel fibre reinforced concrete (SFRC) tunnel segment making it smart. The VWSGs connected to an autonomous wireless node can establish ad-hoc modular networks with other smart segments and the segment properties can be tracked through their whole lifecycle. The main objective pursued are: (i) the design, the implementation, and the performance assessment of the robotic process of installing the sensors; (ii) the design, the fabrication and the mechanical testing of smart segment under cyclical loadings.

This paper explores the potential of satellite Interferometric Synthetic Aperture Radar (InSAR) technology for Structural Health Monitoring (SHM) of road bridges. While many road bridges worldwide are over half a century old and exhibit widespread deterioration, traditional contact-type sensors for SHM are installed only on a few structures, mainly due to their high cost. In recent years, remote sensing techniques, such as satellite InSAR technology, have been explored to overcome these limitations. This paper focuses on the displacements of the Po River Bridge, which is part of the Italian A22 Highway. We extract the bridge’s displacement with Multi-Temporal InSAR data processing using SAR images acquired by the Italian Cosmo-SkyMed mission. We study 8 years of displacement time series of reflective targets, Persistent Scatterers, naturally visible on the bridge without installing any instrumentation on site. We perform an exploratory analysis of the displacements of the entire area through the K-means clustering algorithms and investigate the correlation between the bridge displacements and environmental phenomena (variation of air temperature and river water flow). The results confirm the potential of satellite InSAR technology for the remote monitoring of road bridges and their surrounding area. However, they also highlight the need for a metrological validation of such technology through a direct comparison with measurements from traditional and already validated SHM systems.

In the wind tunnel testing, structural deformation and motion should be measured using non-contact and precisely calibrated techniques. We designed, developed, and set up a novel DIC method for testing large and lightweight structures in wind tunnel. With the use of a universal mount mechanism and optical advancements for far-field structural evaluations, we increased the functionality of our current in-house stereo DIC system. The pitch and azimuth rotations of cameras are taken into consideration during the calibration of the WT-DIC system utilizing the Euler rotation theorem. Case studies for the developed DIC system include a small MAV flapping wing and a large deployable parachute.

As easy-to-deploy, off-the-shelf sensors decrease in cost and increase in accessibility through their integration with well-documented, entry-level electronics platforms, low-quality sensors are increasingly being inappropriately used to characterize physical and natural processes under varying environmental and operational conditions. This is notably occurring across water and hydraulic system applications, which necessitates measuring water levels using ultrasonic sensors. To lay a roadmap for future water and hydraulic system monitoring research and implementation, there is a need to develop a well-informed mapping between sensors and application areas for which the use of the sensors is deemed appropriate. In this work, we identify commonly used ultrasonic sensors, develop systematic experimental setups to simulate their use in common application areas, and evaluate their accuracy under varying conditions and parameters. From the results of these experiments, we present a suggested mapping between sensors and application areas/conditions for which the use of the sensors is appropriate, in addition to limitations placed on each sensor-application pairing are identified.

In recent years, the development of hybrid composites that integrate shape memory alloys (SMAs) into fibrous layered composite materials have become of significant interest. The intrinsic ability of SMAs to change their dynamics and provide energy-dissipation capabilities following phase transformation offers unique opportunities to tune the dynamic of the overall composite. Existing research has focused on the analysis of wire-reinforced composites, while configurations integrating SMA laminae remain to be explored. To bridge this gap, this paper provides a comprehensive study of the dynamic properties of a layered hybrid composite SMA beam under distinct operating conditions. More specifically, the transient response of a hybrid composite beam made of a combination of NiTi SMA and carbon/epoxy is investigated.

Acoustic wave-based devices have attracted greater attention, particularly in the aerospace and bio-medical fields due to their passive and wireless capabilities. Interdigital transducer (IDT) is an integral part of the SHM wave-based sensor, as it transmits information about the structural state. Additionally, embedding the electrodes inside the piezoelectric substrate increases the acoustic coupling and protects the electrodes from potential external damage. This paper uses numerical analysis to discuss sensor responses with different IDT layouts in both frequency and time domains. The results investigate each type’s sensitivity towards mechanical strains and figure of merit, which facilitates the development of an efficient embedded electrode sensor through advanced additive manufacturing techniques.

External lighting is required for autonomous inspections of concrete structures in low-light environments; however, previous studies have only considered uniformly diffused lighting to illuminate images. This study proposes a novel algorithm that utilises angled and directional lighting to obtain pixel-level segmentation of concrete cracks. The method applies a concrete crack detection algorithm to separate images, each illuminated with lighting from a different direction. Using a bitwise OR operation, the findings from all images are combined; the resulting image highlights the extremities of any present cracks in all lighting directions. When tested on a dataset of cracks ranging in widths from 0.07 mm to 0.3 mm, the algorithm obtained recall, precision and F1 score results of 77%, 84% and 92%, respectively. The algorithm was able to correctly segment cracks that were deemed too thin for similar diffused lighting segmentation methods found in literature. The proposed directional lighting algorithm has the potential to improve concrete inspections in low-light environments.

For rapid assessment of infrastructure, the use of minimally invasive sensors that can be deployed remotely using autonomous vehicles is gaining popularity. Such systems are favorable for their ease of deployment and cost-effectiveness. Utilizing electropermanent magnets or adhesives to mount the sensors temporarily forms a barrier between the sensor and the structure being examined. This barrier creates undesirable nonlinearities and transmissibility losses that introduce errors into structural damage detection algorithms. Post-processing of signals using continuous modeling techniques from classical control theory can be applied to the collected signals to remove this error. However, post-processing creates additional analysis steps that require the signal to be taken off device. Processing the data at-the-edge prior to saving it to memory or transmitting it to a base station enables rapid assessment of infrastructure. With minimal time from signal detection to prognostics, such systems can be used in damage forecasting and infrastructure failure prevention. This preliminary work aims to develop a non-linear machine-based compensation technique that is resource and power efficient enough to be processed on-device. The proposed long short-term memory (LSTM) error-compensating network demonstrated potential by increasing the SNRdB by 9.3% and improving RMSE by approximately 20% while widening the usable lower limit of the sensor’s bandwidth from 2.78 to 1.34 Hz. The progress described in this report focuses on setting the framework for the proposed method and paves the way for a full-scale hardware implementation in the near future

This paper presents an integrated system capable of damage imaging of barely visible impact damage (BVID) in composite structures. This system applies guided-wave-based structural health monitoring using 3D digital image correlation, or GWSHM-3D DIC to produce a map of subsurface damage using a short video from a stereo pair of synchronized digital cameras. The proposed system overcomes many limitations of previous efforts of GWSHM that used 2D digital image correlation (DIC). First, 3D DIC can capture the higher-amplitude out-of-plane displacements associated with the anti-symmetric wave mode, lowering the spatial resolution requirements of the cameras. Second, a total wave energy (TWE) imaging condition is employed that uses the monogenic signal via a Reisz transform to obtain the local instantaneous amplitude as a contribution to wave energy. This condition can highlight local resonance in the damage region without the need for high frame rates to fully reconstruct the wavefield. With significantly lowered spatial and temporal resolution requirements of the cameras, high-stiffness materials like composites can be inspected or monitored with a larger field-of-view (FOV). Additionally, signal enhancement techniques intended to increase the effective resolution of the camera are no longer necessary, which reduces the data acquisition time from many hours to a few seconds. To demonstrate this integrated dual-camera concept with the TWE imaging condition, the system was used to image damage in a CFRP composite sandwich panel that had been subjected to a low-velocity impact. Initial damage maps produced for a 100-mm ´ 100-mm FOV using a three-second video pair show precise damage imaging ability that is comparable to benchmark ultrasonic and x-ray scans. This efficient and reliable integrated system demonstrated high potential for in-time damage inspection on composite aircraft and other critical structures.

Reliability assessment of civil structures under seismic loads requires probabilistic evaluation considering the uncertainty of input ground motion and material properties due to deterioration. However, Monte Carlo calculation for the structural reliability analysis is computationally expensive. This study develops the deep kernel learning surrogate model that can not only reduce the computational cost but also provide explainability for the prediction results. The model extracts the features of seismic loads by the convolutional neural network (CNN) and considers the uncertainty of seismic loads and material properties by the Gaussian process regression with the automatic relevance determination (ARD) kernel. By the incorporating gradient-weighted class activation mapping (Grad-CAM) in the CNN, the parts of seismic load response spectra, where contribute to the constructed surrogate model, can be visualized. The model can also provide which input uncertain parameters of structural properties has relatively influence on the output response by the estimated ARD kernel weights. The developed surrogate model is verified by applying it to the seismic performance analysis of a concrete bridge pier with a seismic rubber bearing under various earthquake loads with different intensity and response spectra. The results show that the developed surrogate model can predict accurate distributions of maximum displacements and can provide reasonable contributions of uncertain inputs to enhance the explainability.

Thermal infrared images have been widely employed to detect defections. However, it is challenging to identify the defects from thermal infrared images contaminated by shadows, noise, etc. Those factors do be recorded in the thermal infrared images such that the recorded pixels not only contain the surface temperatures but also have the impacts from the factors. Those factors illustrated in the recorded pixels can be called intensity inhomogeneity. Several researchers have reported that the multiplicative way is feasible to approximate intensity inhomogeneity. A gaussian function is introduced to make sure that the intensity inhomogeneity works on not only a particular pixel but also its neighborhoods contribute. Usually, the fixed window size illustrated in the Gaussian function is used for the whole image for simplification. Intensity inhomogeneity is not spread uniformly over the given image. For those areas with high-intensity inhomogeneity, the Gaussian function with larger window sizes is introduced; otherwise, the Gaussian function with smaller window sizes is supposed to apply to the low-intensity inhomogeneity areas. Adaptive window sizes were proposed such that each pixel will have the Gaussian function with the specified window sizes according to the amount of intensity inhomogeneity. However, the adaptive approach needs a lot of computation, so defect detection will be slowed down. This study proposes the algorithm to modify the algorithm of adaptive window sizes provide an efficient way for defect detection. The proposed approach is based on image entropy; the image entropy will have a bigger value while the intensity inhomogeneity is larger. The image entropy was classified such that limited window sizes were introduced. In doing so, intensity inhomogeneity can be approximated, and the results of image segmentation can identify the defect.

Non-Destructive Evaluation (NDE) methods allow the acquisition of a wide range of information related to the structural performance of bridges and viaducts. Consequently, the outcomes from NDE inspections are often considered in risk assessment procedures to collect detailed information about specific parameters involved in the assessment process. As an example, the defectiveness of bridges and viaducts, usually determined after a thorough visual inspection, commonly plays a key role in assessing the vulnerability of the assets and, therefore, in determining their structural risk conditions. In common practice, the planning of NDE inspections is usually borne by the authority responsible for the bridge inventory according to the prescriptions of a selected Standard and the outcomes from past risk assessments. Due to their recurrence in time and the large number of assets they commonly involve, NDE inspections represent costly operations for management authorities, both in terms of time and economic resources. Considering this, their planning should account for the expenses for their execution in addition to the outcomes from the risk classification of the assets. This paper presents a new NDE bridge inspection prioritization method that considers both risk and cost evaluations to establish inspection priorities within bridge inventories. The information gain criterion is adopted to refine the outcomes from risk assessment and prioritization process. Contemporary, the implementation of the methodology in a GIS framework, together with the use of spatial interpolation methods for data analysis, ensure high interoperability and the rendering of macro-level informative risk maps and inspection plans.

This paper designed and fabricated a distributed fiber optic sensing textile for a composite bridge's structural health monitoring (SHM). Based on the Brillouin optical time-domain analyzer (BOTDA), the sensing textile can achieve the resolution of 1m distributed sensing ability. Unlike electrical sensors, fiber sensing systems enjoy the advantages of resistance to electromagnetic interference, survivability withstanding harsh environments, and can interrogation over kilometers. The embroidery machine from Saint-Gobain embedded the fiber system into the textile material. We have designed a U-shape fiber sensing structure including two arms of 22m each. Each arm can be used as a distributed sensing section. Embedded fiber optic sensing textile would result in reduced installation time, which lowers the labor cost and the work stoppage cost, which can be substantial for certain applications like long-range monitoring. Also, textile provides additional protection and allows the design of different layout patterns to accommodate the requirements of a project. The fiber sensing system was installed inside the girder before the bridge was built. We investigated a novel installation method using slides moving inside the girder and epoxy was applied to fix the sensing textile on the bottom side of the girders. The sensing system was tested after the bridge was built and demonstrated the feasibility of distributed fiber sensing system for monitoring composite bridges. The results indicated the potential of distributed fiber sensing systems in structural health monitoring and provide a solution of small size, low cost, high durability fiber sensing system.

Sensor optimization is an important part of Structural Health Monitoring System (SHM), as the use of different number of modes and the number of sensors have a great impact on the results of sensor optimization, reasonable modality number plays a key role in sensor optimization. But the selection of sensor number and modality mostly depends on experience and economy. In this paper, the method of determining the number of target modes based on modal information incremental matrix of the covariance matrix is studied, and the sensor optimization is carried out by taking a laboratory cable-stayed bridge model as an example, and the results are compared with the method of determining the number of target modes based on the rate of change of the Fisher information matrix. The results show that the method of modal information incremental matrix based on the covariance matrix can accurately calculate the number of target modes required for the optimization of cable-stayed bridge sensors.