Felipe González, Á. Encalada-Dávila, C. Tutivén, B. Puruncajas, Y. Vidal, Carlos Benalcazar-Parra
This work addresses the problem of damage detection on offshore wind turbine jacket-type foundations based on deep learning algorithms. The work utilizes data obtained from the vibration response of a lab-scale wind turbine foundation. The main contributions of this manuscript to damage detection are: (i) an autoencoder neural network trained with only healthy data drawing a normality model, and (ii) a threshold in the function of prediction errors to define the bound limits of damage. The methodology is evaluated using real vibration data from the lab-scale wind turbine foundation tagged with different noise levels and damage scenarios. The results of damage detection show a 100% accuracy, demonstrating that the proposed methodology is practical and promising to be employed in this kind of challenges.
{"title":"DAMAGE DETECTION ON OFFSHORE WIND TURBINE JACKET FOUNDATIONS BASED ON AN AUTOENCODER","authors":"Felipe González, Á. Encalada-Dávila, C. Tutivén, B. Puruncajas, Y. Vidal, Carlos Benalcazar-Parra","doi":"10.12783/shm2021/36264","DOIUrl":"https://doi.org/10.12783/shm2021/36264","url":null,"abstract":"This work addresses the problem of damage detection on offshore wind turbine jacket-type foundations based on deep learning algorithms. The work utilizes data obtained from the vibration response of a lab-scale wind turbine foundation. The main contributions of this manuscript to damage detection are: (i) an autoencoder neural network trained with only healthy data drawing a normality model, and (ii) a threshold in the function of prediction errors to define the bound limits of damage. The methodology is evaluated using real vibration data from the lab-scale wind turbine foundation tagged with different noise levels and damage scenarios. The results of damage detection show a 100% accuracy, demonstrating that the proposed methodology is practical and promising to be employed in this kind of challenges.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125227590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sensor networks facilitate collecting measurement data necessary for decision making regarding structural maintenance and rehabilitation in structural health monitoring (SHM) systems. Nevertheless, the reliability of decision making in SHM systems depends on the proper operation of the sensors. Sensors may exhibit faults, entailing faulty data and incorrect judgment of structural conditions. Therefore, fault diagnosis (FD), comprising detection, isolation, identification, and accommodation of sensor faults, has been introduced in SHM systems, enabling timely detection of faulty data while advancing reliable operation of SHM systems. Traditional FD approaches based on “analytical redundancy” take advantage of correlated sensor data inherent to the SHM system, sometimes neglecting the fault identification step, and are implemented for specific sensor data. In this paper, an analytical redundancy FD approach for SHM systems, coupled with machine learning algorithms and wavelet transforms, capable of processing any type of sensor data is presented. A machine learning (ML) regression algorithm is proposed for fault detection, fault isolation, and fault accommodation, and an ML classification algorithm is proposed for fault identification. Continuous wavelet transform (CWT) is used as a preprocessing step of fault identification, exposing fault patterns in the data. The ML-CWT-FD approach is validated using data from a real-world SHM system in operation at a railway bridge implementing a deep neural network as ML regression algorithm and a convolutional neural network as ML classification algorithm. As a result of this paper, the ML-CWTFD approach is demonstrated to be capable of ensuring reliable SHM systems.
{"title":"SENSOR FAULT DIAGNOSIS COUPLING DEEP LEARNING AND WAVELET TRANSFORMS","authors":"J. J. P. Abadía, H. Fritz, K. Dragos, K. Smarsly","doi":"10.12783/shm2021/36327","DOIUrl":"https://doi.org/10.12783/shm2021/36327","url":null,"abstract":"Sensor networks facilitate collecting measurement data necessary for decision making regarding structural maintenance and rehabilitation in structural health monitoring (SHM) systems. Nevertheless, the reliability of decision making in SHM systems depends on the proper operation of the sensors. Sensors may exhibit faults, entailing faulty data and incorrect judgment of structural conditions. Therefore, fault diagnosis (FD), comprising detection, isolation, identification, and accommodation of sensor faults, has been introduced in SHM systems, enabling timely detection of faulty data while advancing reliable operation of SHM systems. Traditional FD approaches based on “analytical redundancy” take advantage of correlated sensor data inherent to the SHM system, sometimes neglecting the fault identification step, and are implemented for specific sensor data. In this paper, an analytical redundancy FD approach for SHM systems, coupled with machine learning algorithms and wavelet transforms, capable of processing any type of sensor data is presented. A machine learning (ML) regression algorithm is proposed for fault detection, fault isolation, and fault accommodation, and an ML classification algorithm is proposed for fault identification. Continuous wavelet transform (CWT) is used as a preprocessing step of fault identification, exposing fault patterns in the data. The ML-CWT-FD approach is validated using data from a real-world SHM system in operation at a railway bridge implementing a deep neural network as ML regression algorithm and a convolutional neural network as ML classification algorithm. As a result of this paper, the ML-CWTFD approach is demonstrated to be capable of ensuring reliable SHM systems.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127963996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Dardeno, M. Haywood-Alexander, R. Mills, L. Bull, N. Dervilis, K. Worden
Structural health monitoring (SHM) systems have been implemented across multiple engineering applications, and SHM remains an active area of research addressing the improved safety, reliability, and management of these structures. Several challenges, however, have limited the practical implementation and generalisation of SHM technologies, such as operational and environmental fluctuations, repeatability issues, and changes in boundary conditions. These inconsistencies can be problematic for SHM based on machine learning, as healthy states may be incorrectly flagged as damaged, or damaged states may be misclassified as normal variations. Likewise, manufacturing differences can result in variation among similar structures. Accounting for these variations is especially important for a population-based approach to SHM (PBSHM), which seeks to transfer valuable information, including normal operating conditions and damage states, across similar structures. This work aims to quantify this variability, and evaluate the applicability of SHM when these deviations occur. In this paper, an experimental campaign is discussed, in which vibration data were collected over a series of tests on a set of full-scale, composite glider wings. Tests were performed at multiple ambient temperatures, and with real and simulated damage conditions. The frequency response functions of the wings are examined to identify changes in natural frequency.
{"title":"INVESTIGATING THE EFFECTS OF AMBIENT TEMPERATURE ON FEATURE CONSISTENCY IN VIBRATION-BASED SHM","authors":"T. Dardeno, M. Haywood-Alexander, R. Mills, L. Bull, N. Dervilis, K. Worden","doi":"10.12783/shm2021/36344","DOIUrl":"https://doi.org/10.12783/shm2021/36344","url":null,"abstract":"Structural health monitoring (SHM) systems have been implemented across multiple engineering applications, and SHM remains an active area of research addressing the improved safety, reliability, and management of these structures. Several challenges, however, have limited the practical implementation and generalisation of SHM technologies, such as operational and environmental fluctuations, repeatability issues, and changes in boundary conditions. These inconsistencies can be problematic for SHM based on machine learning, as healthy states may be incorrectly flagged as damaged, or damaged states may be misclassified as normal variations. Likewise, manufacturing differences can result in variation among similar structures. Accounting for these variations is especially important for a population-based approach to SHM (PBSHM), which seeks to transfer valuable information, including normal operating conditions and damage states, across similar structures. This work aims to quantify this variability, and evaluate the applicability of SHM when these deviations occur. In this paper, an experimental campaign is discussed, in which vibration data were collected over a series of tests on a set of full-scale, composite glider wings. Tests were performed at multiple ambient temperatures, and with real and simulated damage conditions. The frequency response functions of the wings are examined to identify changes in natural frequency.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130524787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A vibration-based active-sensing global SHM method is proposed and evaluated for its damage localization and quantification accuracy on complex wing structure. In the process, the wing structure is actuated by a white noise vibration and the response signals are collected by a distributed sensor network. The proposed SHM method first utilize auto-regressive exogenous (ARX) model [1] for representing the time-domain response at each sensor location under various damage conditions, where stochasticity contained in structural response is minimized and identified. ARX models are then mapped to damage parameter space via vector-dependent functionally pooled (VFP) method [2]. Then, a damage estimation algorithm based on minimizing VFP-ARX model prediction error is developed. Finally, the damage estimation results are evaluated as the possibility of leveraging multiple senor signal in SHM process is implicated.
{"title":"DAMAGE LOCALIZATION AND MAGNITUDE ESTIMATION ON A COMPOSITE UAV WING VIA STOCHASTIC FUNCTIONALLY POOLED MODELS","authors":"Peiyuan Zhou, Otis Kopsaftopoulos","doi":"10.12783/shm2021/36240","DOIUrl":"https://doi.org/10.12783/shm2021/36240","url":null,"abstract":"A vibration-based active-sensing global SHM method is proposed and evaluated for its damage localization and quantification accuracy on complex wing structure. In the process, the wing structure is actuated by a white noise vibration and the response signals are collected by a distributed sensor network. The proposed SHM method first utilize auto-regressive exogenous (ARX) model [1] for representing the time-domain response at each sensor location under various damage conditions, where stochasticity contained in structural response is minimized and identified. ARX models are then mapped to damage parameter space via vector-dependent functionally pooled (VFP) method [2]. Then, a damage estimation algorithm based on minimizing VFP-ARX model prediction error is developed. Finally, the damage estimation results are evaluated as the possibility of leveraging multiple senor signal in SHM process is implicated.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124291799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Guided wave propagation is a valuable and reliable technique for structural health monitoring (SHM) of aerospace structures. Along with its higher sensitivity towards small damages, it offers advantages in traveling long distances with minimum attenuation. Simulation of guided wave propagation is essential to understand wave behavior, and calculating the dispersion relations forms an integral part of the procedure. Application of the current numerical techniques for complex media is highly involved and faces issues related to accuracy, stability, and computational resources. Development in the field of machine learning and graphical processing units (GPUs) leads to the implementation of a faster, automated, and scalable deep neural networks-based learning approach for such problems. Most of the implementation in the field is based on data collection and uses neural networks for nonlinear mapping from input space to target space. However, a large amount of prior information in the form of a governing differential equation is not utilized. In this paper, we have used Physics-Informed Neural Networks (PINNs), in which neural networks are utilized to solve governing partial differential equations. PINNs are implemented to obtain the solution of a one-dimensional wave equation with Dirichlet boundary conditions. The exact solutions and predicted responses match closely with lower mean square error in limited computational time. We have also conducted a detailed comparison of the effect of neural architecture on the mean square error and the training time. This study shows the merit of deep neural networks leveraging the available physical information to simulate the wave phenomenon for SHM efficiently.
{"title":"SIMULATION OF GUIDED WAVES FOR STRUCTURAL HEALTH MONITORING USING PHYSICS-INFORMED NEURAL NETWORKS","authors":"M. Rautela, M. Raut, S. Gopalakrishnan","doi":"10.12783/shm2021/36297","DOIUrl":"https://doi.org/10.12783/shm2021/36297","url":null,"abstract":"Guided wave propagation is a valuable and reliable technique for structural health monitoring (SHM) of aerospace structures. Along with its higher sensitivity towards small damages, it offers advantages in traveling long distances with minimum attenuation. Simulation of guided wave propagation is essential to understand wave behavior, and calculating the dispersion relations forms an integral part of the procedure. Application of the current numerical techniques for complex media is highly involved and faces issues related to accuracy, stability, and computational resources. Development in the field of machine learning and graphical processing units (GPUs) leads to the implementation of a faster, automated, and scalable deep neural networks-based learning approach for such problems. Most of the implementation in the field is based on data collection and uses neural networks for nonlinear mapping from input space to target space. However, a large amount of prior information in the form of a governing differential equation is not utilized. In this paper, we have used Physics-Informed Neural Networks (PINNs), in which neural networks are utilized to solve governing partial differential equations. PINNs are implemented to obtain the solution of a one-dimensional wave equation with Dirichlet boundary conditions. The exact solutions and predicted responses match closely with lower mean square error in limited computational time. We have also conducted a detailed comparison of the effect of neural architecture on the mean square error and the training time. This study shows the merit of deep neural networks leveraging the available physical information to simulate the wave phenomenon for SHM efficiently.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124944273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yuning Wu, Keping Zhang, Chi-Luen Huang, Sangmin Lee, J. Popovics, Xuan Zhu
Ultrasonic guided waves are of practical interests for nondestructive evaluation (NDE) and structural health monitoring (SHM) since users can promote desirable wave modes for damage detection, thickness measurement, surface condition characterization, stress measurement, and so on. This study focuses on demonstrating the existence of zero group velocity (ZGV) modes for guided waves in free rails. First, the team computed dispersion curves of AREMA standard rails to identify ZGV points through semi-analytical finite element analysis (SAFE). Second, finite element models were established to spatially sample wave propagation in free rails for wavenumberfrequency domain analysis. The results of finite element simulations were compared with dispersion curves produced by SAFE, and multiple points were identified with vanishing group velocity at non-zero wavenumbers. And resonances with positive and negative wavenumbers revealed that the observed standing waves phenomenon results from the interference of two traveling waves propagating with opposite directions. Our observation and developed methodology have potential applications for rail defect detection, support condition assessment, and rail stress measurement.
{"title":"ON THE EXISTENCE OF ZERO GROUP VELOCITY MODES IN RAILS","authors":"Yuning Wu, Keping Zhang, Chi-Luen Huang, Sangmin Lee, J. Popovics, Xuan Zhu","doi":"10.12783/shm2021/36317","DOIUrl":"https://doi.org/10.12783/shm2021/36317","url":null,"abstract":"Ultrasonic guided waves are of practical interests for nondestructive evaluation (NDE) and structural health monitoring (SHM) since users can promote desirable wave modes for damage detection, thickness measurement, surface condition characterization, stress measurement, and so on. This study focuses on demonstrating the existence of zero group velocity (ZGV) modes for guided waves in free rails. First, the team computed dispersion curves of AREMA standard rails to identify ZGV points through semi-analytical finite element analysis (SAFE). Second, finite element models were established to spatially sample wave propagation in free rails for wavenumberfrequency domain analysis. The results of finite element simulations were compared with dispersion curves produced by SAFE, and multiple points were identified with vanishing group velocity at non-zero wavenumbers. And resonances with positive and negative wavenumbers revealed that the observed standing waves phenomenon results from the interference of two traveling waves propagating with opposite directions. Our observation and developed methodology have potential applications for rail defect detection, support condition assessment, and rail stress measurement.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"40 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126123280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pre-stressing is often the decisive feature of concrete bridges. Deterioration of the tendons used to provide pre-stress is thus an immense problem. To detect potential damage, monitoring of the structures promises to be a remedy. However, since the location of damage is seldom known, conventional monitoring systems reach their limits. Then, ultrasound in conjunction with coda wave interferometry becomes promising to monitor entire structures with just a small number of sensors. To demonstrate the general feasibility of detecting tendon rupture, a pre-stressed concrete beam was experimentally investigated. Failure was artificially initiated by cutting a tendon. Ultrasonic and strain measurements recorded stress changes that occurred due to cross-section losses of the tendon. Both the process of pre-stressing and the artificially induced failure has been tracked using the coda wave technique. Based on the relative velocity change of the ultrasound, the internal state change in the (from the outside) still intact specimen could be detected and tracked. These initial results inspire further in-depth investigations into the detection of pre-stressing steel fractures.
{"title":"DAMAGE WITHOUT INDICATION—DETECTION OF TENDON RUPTURE USING CODA WAVE INTERFEROMETRY","authors":"Felix Clauß, M. A. Ahrens, P. Mark","doi":"10.12783/shm2021/36263","DOIUrl":"https://doi.org/10.12783/shm2021/36263","url":null,"abstract":"Pre-stressing is often the decisive feature of concrete bridges. Deterioration of the tendons used to provide pre-stress is thus an immense problem. To detect potential damage, monitoring of the structures promises to be a remedy. However, since the location of damage is seldom known, conventional monitoring systems reach their limits. Then, ultrasound in conjunction with coda wave interferometry becomes promising to monitor entire structures with just a small number of sensors. To demonstrate the general feasibility of detecting tendon rupture, a pre-stressed concrete beam was experimentally investigated. Failure was artificially initiated by cutting a tendon. Ultrasonic and strain measurements recorded stress changes that occurred due to cross-section losses of the tendon. Both the process of pre-stressing and the artificially induced failure has been tracked using the coda wave technique. Based on the relative velocity change of the ultrasound, the internal state change in the (from the outside) still intact specimen could be detected and tracked. These initial results inspire further in-depth investigations into the detection of pre-stressing steel fractures.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125878326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Surface response to excitation (SuRE) and electromechanical impedance methods quantify the difference between the reference and any given spectrums by calculating the sum of the squares of differences (SSD). In part one of this study, twodimensional SSD (2D-SSD) was proposed to quantify the difference of timefrequency plots when the part was excited with the Multiple Width Pulse Excitation (MWPE) signal. In this study, neural networks and deep learning were used for the classification of structural health monitoring (SHM) signals. Since manual encoding of the 2D spectrograms is very complicated to prepare them for classification by using neural networks, deep learning has been used. In this study, the performance of deep learning was evaluated for the classification of sensory data. A cross-shaped part made of PLA was manufactured additively and the center of the part was excited with MWPE and the surface waves were monitored at the end of each extension. Tests were repeated without and with a compressive force at each extension. The recorded time-domain sensory data was converted to spectrogram images using Short-Time Fourier Transform (STFT). The spectrograms were classified with the Convolutional Neural Network (CNN) after proper training. The results showed that the hidden geometry of each extension had a distinctive effect on the characteristics of the monitored signals. CNN could classify the infill type, skin thickness, and loading conditions with better than 92 % accuracy when the responses of the 20 pulses in the MWPE signal were considered.
{"title":"NEW EXCITATION (MULTIPLE WIDTH PULSE EXCITATION (MWPE)) METHOD FOR SHM SYSTEMS—PART 2: CLASSIFICATION OF TIME- FREQUENCY DOMAIN CHARACTERISTICS WITH 2DSSD AND CNN","authors":"Alireza Modir, I. Tansel","doi":"10.12783/shm2021/36345","DOIUrl":"https://doi.org/10.12783/shm2021/36345","url":null,"abstract":"Surface response to excitation (SuRE) and electromechanical impedance methods quantify the difference between the reference and any given spectrums by calculating the sum of the squares of differences (SSD). In part one of this study, twodimensional SSD (2D-SSD) was proposed to quantify the difference of timefrequency plots when the part was excited with the Multiple Width Pulse Excitation (MWPE) signal. In this study, neural networks and deep learning were used for the classification of structural health monitoring (SHM) signals. Since manual encoding of the 2D spectrograms is very complicated to prepare them for classification by using neural networks, deep learning has been used. In this study, the performance of deep learning was evaluated for the classification of sensory data. A cross-shaped part made of PLA was manufactured additively and the center of the part was excited with MWPE and the surface waves were monitored at the end of each extension. Tests were repeated without and with a compressive force at each extension. The recorded time-domain sensory data was converted to spectrogram images using Short-Time Fourier Transform (STFT). The spectrograms were classified with the Convolutional Neural Network (CNN) after proper training. The results showed that the hidden geometry of each extension had a distinctive effect on the characteristics of the monitored signals. CNN could classify the infill type, skin thickness, and loading conditions with better than 92 % accuracy when the responses of the 20 pulses in the MWPE signal were considered.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114832202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Prasad Chetti, Hesham Ali, D. Ghersi, R. Gandhi, Brian Ricks, Lotfollah Najjar
Public safety and economic growth are some of the key factors in motivating governments to keep their civil infrastructures, in particular bridges, safe and sound. However, the American Society for Civil Engineers gave a C+ grade for U.S. bridges in 2017. It has been observed that many parameters associated with bridges, such as geographical locations, designs, materials used, and traffic patterns, play key roles in determining the safety and deterioration rates of bridges. However, there is still a lack of studies that analyze the exact impact of all relevant parameters. The motivation of this study is to propose a new data-driven model that employs the concept of population analysis in assessing the impact of each potential parameter and extracting critical information associated with civil infrastructures and their deterioration patterns. We use a correlation network model to analyze and visualize the big data associated with more than 600,000 bridges in the national bridge inventory database. Graph theoretic analysis is applied to the correlation networks to find elements or clusters of interest. A sub-set of 268 bridges across the US of the same age are considered for this case study and the Markov clustering algorithm is used to obtain the clusters from the correlation network. Enrichment analysis is applied to the clusters to identify the significantly enriched input parameters. Preliminary results reveal several facts, including that prestressed concrete bridges in the Southeast region perform better than steel bridges in the Midwestern region. The obtained results are supported by previous research and further validated by the exploratory factor analysis when dividing the clusters into two groups. The proposed network model provides a new data-driven methodology for evaluating the safety and performance of structures. It provides domain experts with valuable information on how to efficiently allocate time and funds for inspecting existing bridges and how to identify key bridge parameters suitable for designing and constructing new bridges in various geographical areas.
{"title":"A NEW APPROACH FOR ANALYZING SAFETY AND PERFORMANCE FACTORS IN CIVIL INFRASTRUCTURES USING CORRELATION NETWORKS AND POPULATION ANALYSIS","authors":"Prasad Chetti, Hesham Ali, D. Ghersi, R. Gandhi, Brian Ricks, Lotfollah Najjar","doi":"10.12783/shm2021/36305","DOIUrl":"https://doi.org/10.12783/shm2021/36305","url":null,"abstract":"Public safety and economic growth are some of the key factors in motivating governments to keep their civil infrastructures, in particular bridges, safe and sound. However, the American Society for Civil Engineers gave a C+ grade for U.S. bridges in 2017. It has been observed that many parameters associated with bridges, such as geographical locations, designs, materials used, and traffic patterns, play key roles in determining the safety and deterioration rates of bridges. However, there is still a lack of studies that analyze the exact impact of all relevant parameters. The motivation of this study is to propose a new data-driven model that employs the concept of population analysis in assessing the impact of each potential parameter and extracting critical information associated with civil infrastructures and their deterioration patterns. We use a correlation network model to analyze and visualize the big data associated with more than 600,000 bridges in the national bridge inventory database. Graph theoretic analysis is applied to the correlation networks to find elements or clusters of interest. A sub-set of 268 bridges across the US of the same age are considered for this case study and the Markov clustering algorithm is used to obtain the clusters from the correlation network. Enrichment analysis is applied to the clusters to identify the significantly enriched input parameters. Preliminary results reveal several facts, including that prestressed concrete bridges in the Southeast region perform better than steel bridges in the Midwestern region. The obtained results are supported by previous research and further validated by the exploratory factor analysis when dividing the clusters into two groups. The proposed network model provides a new data-driven methodology for evaluating the safety and performance of structures. It provides domain experts with valuable information on how to efficiently allocate time and funds for inspecting existing bridges and how to identify key bridge parameters suitable for designing and constructing new bridges in various geographical areas.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127007200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alexander Thoms, Gabriel Earle, Nicholas Charron, Sven Malama, S. Narasimhan
Advances in robotic mapping, planning, and perception have spurred applications-based robotics research in the domain of infrastructure inspection and preservation. Though a significant portion of this research has centered around the use of unmanned aerial, ground, and underwater vehicles, research in the use of unmanned surface vehicles (USVs) is limited. USVs present a unique opportunity to capture combined maps above and below water, which is essential for the inspection of waterspanning bridges, harbors, dams, and levees. In this paper, we investigate the use of USVs for infrastructure inspection by outfitting a USV platform with a multibeam sonar, horizontally and vertically mounted lidars, several ruggedized RGB cameras, and a high-rate inertial measurement unit (IMU). By time-synchronizing all sensors, we are able to fuse information collected from lidar, camera, and IMU units via tightly-coupled lidar-visual-inertial (LVI) simultaneous mapping and localization (SLAM). We validate our methodology by collecting sensory data of an abandoned quarry and by generating a combined 3D point cloud map using lidar data, multibeam sonar data, and maximum a posteriori trajectory from the LVI SLAM approach. Experiments validate the performance of the proposed USV system, highlighting challenges in extrinsic calibration of non-overlapping sensors, sonar denoising, and refined inter-keyframe pose estimation for key-frame based SLAM approaches.
{"title":"COMBINED LIDAR AND SONAR MAPPING FOR PARTIALLY SUBMERGED INFRASTRUCTURE","authors":"Alexander Thoms, Gabriel Earle, Nicholas Charron, Sven Malama, S. Narasimhan","doi":"10.12783/shm2021/36336","DOIUrl":"https://doi.org/10.12783/shm2021/36336","url":null,"abstract":"Advances in robotic mapping, planning, and perception have spurred applications-based robotics research in the domain of infrastructure inspection and preservation. Though a significant portion of this research has centered around the use of unmanned aerial, ground, and underwater vehicles, research in the use of unmanned surface vehicles (USVs) is limited. USVs present a unique opportunity to capture combined maps above and below water, which is essential for the inspection of waterspanning bridges, harbors, dams, and levees. In this paper, we investigate the use of USVs for infrastructure inspection by outfitting a USV platform with a multibeam sonar, horizontally and vertically mounted lidars, several ruggedized RGB cameras, and a high-rate inertial measurement unit (IMU). By time-synchronizing all sensors, we are able to fuse information collected from lidar, camera, and IMU units via tightly-coupled lidar-visual-inertial (LVI) simultaneous mapping and localization (SLAM). We validate our methodology by collecting sensory data of an abandoned quarry and by generating a combined 3D point cloud map using lidar data, multibeam sonar data, and maximum a posteriori trajectory from the LVI SLAM approach. Experiments validate the performance of the proposed USV system, highlighting challenges in extrinsic calibration of non-overlapping sensors, sonar denoising, and refined inter-keyframe pose estimation for key-frame based SLAM approaches.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"2006 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125610921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：