This study presentes a new nonparametric structural damage detection algorithm that integrates self-organizing maps with a pattern-recognition neural network to quantify and locate structural damage. In this algorithm, self-organizing maps are used to extract a number of damage indices from the ambient vibration response of the monitored structure. The presented study is unique because it demonstrates the development of a nonparametric vibration-based damage detection algorithm that utilizes self-organizing maps to extract meaningful damage indices from ambient vibration signals in the time domain. The ability of the algorithm to identify damage was demonstrated analytically using a finite-element model of a hot-rolled steel grid structure. The algorithm successfully located the structural damage under several damage cases, including damage resulting from local stiffness loss in members and damage resulting from changes in boundary conditions. A sensitivity study was also conducted to evaluate the effects of noise on the computed damage indices. The algorithm was proved to be successful even when the signals are noise-contaminated.
One of the main tasks in structural health monitoring process is to create reliable algorithms that are capable of translating the measured response into meaningful information reflecting the actual condition of the monitored structure. The authors have recently introduced a novel unsupervised vibration-based damage detection algorithm that utilizes self-organizing maps to quantify structural damage and assess the overall condition of structures. Previously, this algorithm had been tested using the experimental data of Phase II Experimental Benchmark Problem of Structural Health Monitoring, introduced by the IASC (International Association for Structural Control) and ASCE (American Society of Civil Engineers). In this paper, the ability of this algorithm to quantify structural damage is tested analytically using an experimentally validated finite element model of a laboratory structure constructed at Qatar University.
The study presented in this paper introduces a new intelligent methodology to mitigate the vibration response of flexible cantilever plates. The use of the piezoelectric sensor/actuator pairs for active control of plates is discussed. An intelligent neural network based controller is designed to control the optimal voltage applied on the piezoelectric patches. The control technique utilizes a neurocontroller along with a Kalman Filter to compute the appropriate actuator command. The neurocontroller is trained based on an algorithm that incorporates a set of emulator neural networks which are also trained to predict the future response of the cantilever plate. Then, the neurocontroller is evaluated by comparing the uncontrolled and controlled responses under several types of dynamic excitations. It is observed that the neurocontroller reduced the vibration response of the flexible cantilever plate significantly; the results demonstrated the success and robustness of the neurocontroller independent of the type and distribution of the excitation force.
Metamaterial inspired structures, or metastructures, are structural members that incorporate periodic or non-periodic inserts. Recently, a new class of metastructures has been introduced which feature chiral lattice inserts. It was found that this type of inserts has frequency bandgaps which can be tuned by altering the geometry of the chiral lattice. Previous studies have shown that inserting non-periodic chiral lattices inside a beam-like structure results in efficient vibration attenuation at low frequencies. In the study presented in this paper, a genetic algorithm based optimization technique is developed to automatically generate chiral lattices which are tuned to suppress vibration in a flexible beam-like structure. Several parameters are incorporated in the optimization process such as the radius of circular nodes and characteristic angle as well as the spacing and distribution of circular inserts. The efficiency of the …
One of the major challenges in civil, mechanical, and aerospace engineering is to develop vibration suppression systems with high efficiency and low cost. Recent studies have shown that high damping performance at broadband frequencies can be achieved by incorporating periodic inserts with tunable dynamic properties as internal resonators in structural systems. Structures featuring these kinds of inserts are referred to as metamaterials inspired structures or metastructures. Chiral lattice inserts exhibit unique characteristics such as frequency bandgaps which can be tuned by varying the parameters that define the lattice topology. Recent analytical and experimental investigations have shown that broadband vibration attenuation can be achieved by including chiral lattices as internal resonators in beam-like structures. However, these studies have suggested that the performance of chiral lattice inserts can be maximized by utilizing an efficient optimization technique to obtain the optimal topology of the inserted lattice. In this study, an automated optimization procedure based on a genetic algorithm is applied to obtain the optimal set of parameters that will result in chiral lattice inserts tuned properly to reduce the global vibration levels of a finite-sized beam. Genetic algorithms are considered in this study due to their capability of dealing with complex and insufficiently understood optimization problems. In the optimization process, the basic parameters that govern the geometry of periodic chiral lattices including the number of circular nodes, the thickness of the ligaments, and the characteristic angle are considered. Additionally, a new set of parameters is introduced to enable the optimization process to explore non-periodic chiral designs. Numerical simulations are carried out to demonstrate the efficiency of the optimization process.
Structural damage detection has been an interdisciplinary area of interest for various engineering fields. While the available damage detection methods have been in the process of adapting machine learning concepts, most machine learning based methods extract “hand-crafted” features which are fixed and manually selected in advance. Their performance varies significantly among various patterns of data depending on the particular structure under analysis. Convolutional neural networks (CNNs), on the other hand, can fuse and simultaneously optimize two major sets of an assessment task (feature extraction and classification) into a single learning block during the training phase. This ability not only provides an improved classification performance but also yields a superior computational efficiency. 1D CNNs have recently achieved state-of-the-art performance in vibration-based structural damage detection; however, it has been reported that the training of the CNNs requires significant amount of measurements especially in large structures. In order to overcome this limitation, this paper presents an enhanced CNN-based approach that requires only two measurement sets regardless of the size of the structure. This approach is verified using the experimental data of the Phase II benchmark problem of structural health monitoring which had been introduced by IASC-ASCE Structural Health Monitoring Task Group. As a result, it is shown that the enhanced CNN-based approach successfully estimated the actual amount of damage for the nine damage scenarios of the benchmark study.
Structural health monitoring (SHM) and vibration-based structural damage detection have been a continuous interest for civil, mechanical and aerospace engineers over the decades. Early and meticulous damage detection has always been one of the principal objectives of SHM applications. The performance of a classical damage detection system predominantly depends on the choice of the features and the classifier. While the fixed and hand-crafted features may either be a sub-optimal choice for a particular structure or fail to achieve the same level of performance on another structure, they usually require a large computation power which may hinder their usage for real-time structural damage detection. This paper presents a novel, fast and accurate structural damage detection system using 1D Convolutional Neural Networks (CNNs) that has an inherent adaptive design to fuse both feature extraction and classification blocks into a single and compact learning body. The proposed method performs vibration-based damage detection and localization of the damage in real-time. The advantage of this approach is its ability to extract optimal damage-sensitive features automatically from the raw acceleration signals. Large-scale experiments conducted on a grandstand simulator revealed an outstanding performance and verified the computational efficiency of the proposed real-time damage detection method.
Vibration suppression remains a crucial issue in the design of structures and machines. Recent studies have shown that with the use of metamaterial inspired structures (or metastructures), considerable vibration attenuation can be achieved. Optimization of the internal geometry of metastructures maximizes the suppression performance. Zigzag inserts have been reported to be efficient for vibration attenuation. It has also been reported that the geometric parameters of the inserts affect the vibration suppression performance in a complex manner. In an attempt to find out the most efficient parameters, an optimization study has been conducted on the linear zigzag inserts and is presented here. The research reported in this paper aims at developing an automated method for determining the geometry of zigzag inserts through optimization. This genetic algorithm based optimization process searches for optimal zigzag designs which are properly tuned to suppress vibrations when inserted in a specific host structure (cantilever beam). The inserts adopted in this study consist of a cantilever zigzag structure with a mass attached to its unsupported tip. Numerical simulations are carried out to demonstrate the efficiency of the proposed zigzag optimization approach.
Operational modal analysis, OMA, results in unscaled mode shapes, since no forces are measured. Yet, obtaining a scaled modal model, i.e. knowing the modal mass of each mode (assuming proportional damping), is essential in many cases for structural health monitoring and load estimation. Several methods have therefore recently been developed for this purpose. The so-called OMAH method is a recently developed method for scaling OMA models, based on harmonic excitation of the structure. A number of frequencies are excited, one by one, and for each frequency, one or more frequency response values are calculated, that are then used for estimation of the modal masses of each mode, and residual effects of modes outside the frequency of interest. In the present paper, measurements were made on a four-story office building which was excited with a small, 200 N sine peak electrodynamic shaker. It is demonstrated that this small shaker was sufficient to excite the building with a force level of approx.. 1.8 N RMS close to the first eigenfrequency of the building, which was sufficient to produce harmonic response across the building. Reliable modal masses were possible to obtain within an accuracy of 6%. This demonstrates the feasibility of the OMAH method.
In softwood species, annual ring width correlates with various timber characteristics, including the density and modulus of elasticity along with bending and tensile strengths. Knowledge of annual ring profiles may contribute to more accurate machine strength grading of sawn timber. This paper proposes a fast and accurate method for automatic estimation of ring profiles along timber boards on the basis of optical scanning. The method utilizes two 1D convolutional neural networks to determine the pith location and detect the surface annual rings at multiple cross-sections along the scanned board. The automatically extracted rings and pith information can then be used to estimate the annual ring profile at each cross-section. The proposed method was validated on a large number of board cross-sections for which the pith locations and radial ring width profiles had been determined manually. The paper also investigates the potential of using the automatically estimated average ring width as an indicating property in machine strength grading of sawn timber. The results indicated that combining the automatically estimated ring width with other prediction variables can improve the accuracy of bending and tensile strength predictions, especially when the grading is based only on information extracted from optical and laser scanning data.(C) 2022 The Author(s). Published by Elsevier Ltd.
Walking-induced loads on office floors can generate unwanted vibrations. The current multi-person loading models are limited since they do not take into account nondeterministic factors such as pacing rates, walking paths, obstacles in walking paths, busyness of floors, stride lengths, and interactions among the occupants. This study proposes a novel video-vibration monitoring system to investigate the complex human walking patterns on floors. The system is capable of capturing occupant movements on the floor with cameras, and extracting walking trajectories using image processing techniques. To demonstrate its capabilities, the system was installed on a real office floor and resulting trajectories were statistically analyzed to identify the actual walking patterns, paths, pacing rates, and busyness of the floor with respect to time. The correlation between the vibration levels measured by the wireless sensors and the trajectories extracted from the video recordings were also investigated. The results showed that the proposed video-vibration monitoring system has strong potential to be used in training data-driven crowd models, which can be used in future studies to generate realistic multi-person loading scenarios.
This paper presents a fast, accurate, and simple systematic approach for online condition monitoring and severity identification of ball bearings. This approach utilizes compact one-dimensional (1-D) convolutional neural networks (CNNs) to identify, quantify, and localize bearing damage. The proposed approach is verified experimentally under several single and multiple damage scenarios. The experimental results demonstrated that the proposed approach can achieve a high level of accuracy for damage detection, localization, and quantification. Besides its real-time processing ability and superior robustness against the high-level noise presence, the compact and minimally trained 1-D CNNs in the core of the proposed approach can handle new damage scenarios with utmost accuracy.
Internationally, an annual number of more than a million fatalities are caused by road traffic crashes, with particularly signalized intersections being crash prone locations within the highway system. An accumulation of conflicts between drivers is caused by the different movements (through and turning) from different directions at the intersection; hence, studying the trajectories of turning vehicles is an important step towards improving traffic safety performance of these facilities. In view of that, the current paper aims at providing further insight into the behaviour of left-turning vehicles (right-hand traffic rule) at signalized intersections in the State of Qatar. At first, a total of 44 trajectories of free-flowing vehicles were manually extracted from a recorded video for a single approach of Lekhwair signalized intersection in Doha City, State of Qatar. After that, the extracted trajectories were statistically analysed in an attempt to explore the factors affecting the path of left-turning vehicles at signalized intersections. The results suggest that the characteristics of the extracted paths are significantly related to the vehicle’s entry speed, minimum speed throughout its turning manoeuvre, and the lateral distance between the exit point and the curb (i.e., targeted exit lane). Provided that the speed parameters can be fairly an indication to the driving behaviour, it can be concluded that the driver’s attitude plays an important role in drawing the manoeuvre of a turning vehicle as does the pre-selection of the exit lane. Finally, the effort presented in this paper can be regarded as a way forward towards understanding the behaviour of turning vehicles at signalised intersection in the State of Qatar.
This paper aims at developing a convolutional neural network (CNN)-based tool that can automatically detect the left-turning vehicles (right-hand traffic rule) at signalized intersections and extract their trajectories from a recorded video. The proposed tool uses a region-based CNN trained over a limited number of video frames to detect moving vehicles. Kalman filters are then used to track the detected vehicles and extract their trajectories. The proposed tool achieved an acceptable accuracy level when verified against the manually extracted trajectories, with an average error of 16.5 cm. Furthermore, the trajectories extracted using the proposed vehicle tracking method were used to demonstrate the applicability of the minimum-jerk principle to reproduce variations in the vehicles’ paths. The effort presented in this paper can be regarded as a way forward toward maximizing the potential use of deep learning in traffic safety applications.
Studies with large physical models are a vital link between the theoretical work and field applications provided that these models are designed to represent real structures where various types and levels of uncertainties can be incorporated. While comprehensive analytical and laboratory joint studies are ongoing at Qatar University, University of Central Florida and University of Alberta, this paper presents the initial findings of dynamic testing at Qatar University. A laboratory stadium structure (grandstand simulator) has been constructed at Qatar University. Capable of housing thirty spectators, Qatar University grandstand simulator is arguably the largest laboratory stadium in the world. The structure is designed in a way that several different structural configurations can be tested in laboratory conditions to enable researchers to test newly developed damage detection algorithms. The study presented in this paper covers the finite element modeling and modal testing of the test structure.
Floating-Slab Tracks (FST) are predominantly used for mitigating railway-induced vibrations where the concrete slab is mounted on soft resilient bearings to provide vibration isolation. This paper presents a research study on the dynamic behavior of a special type of FST used in the recently built subway system in Doha, Qatar. The special FST has a continuous concrete slab with periodic grooves. Therefore, the track can be modeled as a periodic structure with a slab unit having two elements with different cross-sections. Extensive numerical and experimental investigations were conducted on a multi-unit full-scale mockup track representing the special FST. A fast running model based on the Dynamic Stiffness Method was developed and examined, in an initial numerical exercise, against a detailed Finite Element model for a track with a finite length. In the experimental campaign, a test was performed with an impact hammer to identify the actual vibration response of the mockup track. Results from the experimental investigations were then used for model updating of the fast running model. The model updating process was carried out according to an automated hybrid optimization approach that combines genetic algorithms with a local search method. Finally, the updated model was extended to an infinite model to investigate the influence of varying grooves thickness on the dynamic behavior of the special track with infinite length for both bending and torsion scenarios. The investigations suggested that reducing the thickness below 50% of the full thickness of the slab significantly affects the dynamic behavior of the special FST.
Evaluating the severity of structural damage is a critical component of Structural Health Monitoring (SHM). Convolutional Neural Networks (CNNs) have been used before to detect structural damage and evaluate its severity by utilising only raw vibration data. However, these vibration-based CNN applications were limited to discrete user-defined levels of damage. To provide a more accurate representation of structural damage, this paper aims to design and validate a framework for evaluating structural damage severity within a continuous range of damage levels, using 1D CNNs and distributed raw acceleration data. To this purpose, a simple Finite Element (FE) cantilever model with non-rigid rotational spring support was adopted. Damage was simulated at the support as reduction of the rotational spring stiffness. The performance of the proposed framework was assessed under different excitation scenarios and data pre-processing techniques. The results demonstrate the ability of 1D CNNs to evaluate damage severity with high accuracy. By estimating the reduced value of the rotational spring stiffness, the proposed framework can also be used towards FE model updating in parallel with damage severity evaluation.
The study presented in this paper is arguably the first study to use a self-organizing map (SOM) for global structural damage detection. A novel unsupervised vibration-based damage detection algorithm is introduced using SOMs in order to quantify structural damage. In this algorithm, SOMs are used to extract a number of damage indices from the random acceleration response of the monitored structure in the time domain. The summation of the indices is used as an indicator which reflects the overall condition of the structure. The ability of the algorithm to quantify the overall structural damage is demonstrated using experimental data of Phase II experimental benchmark problem of structural health monitoring.
The stadium structures have unique structural features increasing the significance of structural monitoring systems specifically designed for them. Aside from vibrations serviceability concerns and human -induced excitations, the development and propagation of structural damage under all possible atmospheric and seismic conditions need to be closely monitored for structural resiliency and integrity of the stadia. As such, Structural Health Monitoring (SHM) methods combined with effective data evaluation methodologies need to be deployed to monitor the structural performance of stadiums. Even though stadia monitoring has been performed at multiple locations in the world, a web based and real-time SHM network of stadia is not known to authors. As a preliminary study for the network implementation of stadia monitoring with acceleration measurements, the presented work focuses on the fundamental steps to accomplish this goal, with a collaborative research effort between Qatar University, the University of Central Florida, and University of Alberta. The authors performed analytical investigations and experimental testing on stadium -type structures built in laboratory environments for the development of the SHM framework. Specialized signal processing algorithms, sensing suites and approaches considering multi -scale monitoring were used on collected acceleration measurements. The novelty of the work presented in this manuscript are the following items which exist simultaneously in the developed SHM framework. The developed framework is a web -based monitoring application where structural damage is detected in real-time. The proposed methodology operates directly on raw acceleration signals and runs at a network level. With that, the damage detection, damage localization, and damage quantification tasks are performed simultaneously, while the feature extraction and classification stages are combined in one learning body.
This paper presents a brief overview of vibration-based damage identification studies based on Deep Learning (DL) in civil engineering structures. The presence, type, size, and propagation of structural damage on civil infrastructure have always been a topic of research. In the last couple of decades, there has been a significant shift in the damage detection paradigm when the advancements in sensing and computing technologies met with the ever-expanding use of artificial neural network algorithms. Machine-Learning (ML) tools enabled researchers to implement more feasible and faster tools in damage detection applications. When an artificial neural network has more than three layers, it is typically considered as a âdeepâ learning network. Being an important accomplishment of the ML era, DL tools enable complex systems which are made of several layers to learn implementations of data with outstanding categorization and compartmentalization capability. In fact, with proper training, a DL tool can operate directly with the unprocessed raw data and help the algorithm produce output data. Competitive capabilities like this led DL algorithms perform very well in complicated problems by dividing a relatively large problem into much smaller and more manageable portions. Specifically for damage identification and localization on civil infrastructure, Convolutional Neural Networks (CNNs) and Unsupervised Pretrained Networks (UPNs) are the known DL tools published in the literature. This paper presents an overview of these studies. © 2022, The Society for Experimental Mechanics, Inc.
This paper presents a brief overview of vibration-based structural damage detection studies that are based on machine learning (ML) in civil engineering structures. The review includes both parametric and nonparametric applications of ML accompanied with analytical and/or experimental studies. While the ML tools help the system learn from the data fed into, the computer enhances the task with the learned information without any programming on how to process the relevant data. As such, the performance level of ML-based damage identification methodologies depends on the feature extraction and classification steps, especially on the classifier choices for which the characteristic nature of the acceleration signals is recorded in a feasible way. Yet, there are several issues to be discussed about the existing ML procedures for both parametric and nonparametric applications, which are presented in this paper.
In this paper, a novel one dimensional convolution neural network (1D-CNN) based structural damage assessment technique is validated with a benchmark study published by IASC-ASCE Structural Health Monitoring Task Group in 2003. In contrast with predominant machine learning based structural damage detection techniques of the literature, the technique shown in this paper runs without manual feature extraction or preprocessing stages. It runs directly on the raw vibration data. In CNNs, the stages of feature extraction and feature classification are merged into one stage; therefore, the proposed technique is efficient, feasible and economical. Utilizing the optimal features learned by 1D CNNs, the proposed CNN-based technique considerably improves the classification efficiency and accuracy. The performance improvement of the proposed technique is assessed by calculating the “Probability of Damage” values for damage estimations. The unseen structural damage cases between the two extreme end structural cases (zero damage and total damage) were successfully identified. Consequently, it is validated that the improved CNN-based technique is efficient since it predicted the level of damage consistently with the structural damage cases defined in the existing benchmark.
Monitoring the structural performance of engineering structures has always been pertinent for maintaining structural health and assessing the life cycle of structures. Structural Health Monitoring (SHM) and Structural Damage Detection (SDD) fields have been topics of ongoing research over the years to explore and verify different monitoring techniques and damage detection and localization procedures. In an attempt to compare performances of different methods, benchmark datasets are valuable resources since the data is made available to researchers enabling side-by-side comparisons. This paper presents a new experimental benchmark dataset generated from tests on a large-scale laboratory structure. The primary goal of the authors was to explore brand-new damage detection and quantification methodologies for efficient monitoring of structures. For this purpose, a large-scale steel grid structure with footprint dimensions of 4.2 m Ã 4.2 m was constructed in laboratory environment and it has been used as a test bed by the authors. The structural members of the structure are all IPE120 hot-rolled steel cross sections. The simulation of structural damage was simply loosening the bolts at one of the beam-to-girder connections, which is a slight change of rotational stiffness at the joint of the steel grid structure. The authors shared the dataset for 1 undamaged and 30 damaged conditions and published it on a public website as a new benchmark problem for structural damage detection at http://www.structuralvibration.com/benchmark/ so that other researchers can use the data and test algorithms. The authors also shared one of the damage detection tools they used, One-Dimensional Convolutional Neural Networks (1D-CNNs). The application codes, configuration files, and accompanied components of the 1D-CNNs package are available for viewers at http://www.structuralvibration.com/cnns/. © 2022, The Society for Experimental Mechanics, Inc.
Monitoring structural damage is extremely important for sustaining and preserving the service life of civil structures. While successful monitoring provides resolute and staunch information on the health, serviceability, integrity and safety of structures; maintaining continuous performance of a structure depends highly on monitoring the occurrence, formation and propagation of damage. Damage may accumulate on structures due to different environmental and human-induced factors. Numerous monitoring and detection approaches have been developed to provide practical means for early warning against structural damage or any type of anomaly. Considerable effort has been put into vibration-based methods, which utilize the vibration response of the monitored structure to assess its condition and identify structural damage. Meanwhile, with emerging computing power and sensing technology in the last decade, Machine Learning (ML) and especially Deep Learning (DL) algorithms have become more feasible and extensively used in vibration-based structural damage detection with elegant performance and often with rigorous accuracy. While there have been multiple review studies published on vibration-based structural damage detection, there has not been a study where the transition from traditional methods to ML and DL methods are described and discussed. This paper aims to fulfill this gap by presenting the highlights of the traditional methods and provide a comprehensive review of the most recent applications of ML and DL algorithms utilized for vibration-based structural damage detection in civil structures.
Being an alternative to conventional wired sensors, wireless sensor networks (WSNs) are extensively used in Structural Health Monitoring (SHM) applications. Most of the Structural Damage Detection (SDD) approaches available in the SHM literature are centralized as they require transferring data from all sensors within the network to a single processing unit to evaluate the structural condition. These methods are found predominantly feasible for wired SHM systems; however, transmission and synchronization of huge data sets in WSNs has been found to be arduous. As such, the application of centralized methods with WSNs has been a challenge for engineers. In this paper, the authors are presenting a novel application of 1D Convolutional Neural Networks (1D CNNs) on WSNs for SDD purposes. The SDD is successfully performed completely wireless and real-time under ambient conditions. As a result of this, a decentralized damage detection method suitable for wireless SHM systems is proposed. The proposed method is based on 1D CNNs and it involves training an individual 1D CNN for each wireless sensor in the network in a format where each CNN is assigned to process the locally-available data only, eliminating the need for data transmission and synchronization. The proposed damage detection method operates directly on the raw ambient vibration condition signals without any filtering or preprocessing. Moreover, the proposed approach requires minimal computational time and power since 1D CNNs merge both feature extraction and classification tasks into a single learning block. This ability is prevailingly cost-effective and evidently practical in WSNs considering the hardware systems have been occasionally reported to suffer from limited power supply in these networks. To display the capability and verify the success of the proposed method, large-scale experiments conducted on a laboratory structure equipped with a state-of-the-art WSN are reported.
This paper presents a method for active vibration control of smart thin cantilever plates. For model formulation needed for controller design and simulations, finite difference technique is used on the cantilever plate response calculations. Piezoelectric patches are used on the plate, for which a neural network based control algorithm is formed and a neurocontroller is produced to calculate the required voltage to be applied on the actuator patch. The neurocontroller is trained and run with a Kalman Filter for controlling the structural response. The neurocontroller performance is assessed by comparing the controlled and uncontrolled structural responses when the plate is subjected to various excitations. It is shown that the acceleration response of the cantilever plate is suppressed considerably validating the efficacy of the neurocontroller and the success of the proposed methodology.
Structural damage detection methods available for structural health monitoring applications are based on data preprocessing, feature extraction, and feature classification. The feature classification task requires considerable computational power which makes the utilization of centralized techniques relatively infeasible for wireless sensor networks. In this paper, the authors present a novel Wireless Sensor Network (WSN) based on One Dimensional Convolutional Neural Networks (1D CNNs) for real-time and wireless structural health monitoring (SHM). In this method, each CNN is assigned to its local sensor data only and a corresponding 1D CNN is trained for each sensor unit without any synchronization or data transmission. This results in a decentralized system for structural damage detection under ambient environment. The performance of this method is tested and validated on a steel grid laboratory structure.
Most of the classical structural damage detection systems involve two processes, feature extraction and feature classification. Usually, the feature extraction process requires large computational effort which prevent the application of the classical methods in real-time structural health monitoring applications. Furthermore, in many cases, the hand-crafted features extracted by the classical methods fail to accurately characterize the acquired signal, resulting in poor classification performance. In an attempt to overcome these issues, this paper presents a novel, fast and accurate structural damage detection and localization system utilizing one dimensional convolutional neural networks (CNNs) arguably for the first time in SHM applications. The proposed method is capable of extracting optimal damage-sensitive features automatically from the raw acceleration signals, allowing it to be used for real-time damage detection. This paper presents the preliminary experiments conducted to verify the proposed CNN-based approach.
The use of self-organizing maps and artificial neural networks for structural health monitoring is presented in this paper. The authors recently developed a nonparametric structural damage detection algorithm for extracting damage indices from the ambient vibration response of a structure. The algorithm is based on self-organizing maps with a multilayer feedforward pattern recognition neural network. After the training of the self-organizing maps, the algorithm was tested analytically under various damage scenarios based on stiffness reduction of beam members and boundary condition changes of a grid structure. The results indicated that proposed algorithm can successfully locate and quantify damage on the structure.
Metastructures are known to provide considerable vibration attenuation for mechanical systems. With the optimization of the internal geometry of metastructures, the suppression performance of the host structure increases. While the zigzag inserts have been shown to be efficient for vibration attenuation, the geometric properties of the inserts affect the suppression performance in a complex manner when attached to the host structure. This paper presents a genetic algorithm based optimization study conducted to come up with the most efficient geometric properties of the zigzag inserts. The inserts studied in this paper are simply cantilever zigzag structures with a mass attached to the unsupported tips. Numerical simulations are run to show the efficiency of the optimization process.
This paper presents a novel real-time rotating machinery damage monitoring system. The system detects, quantifies, and localizes damage in ball bearings in a fast and accurate way using one-dimensional convolutional neural networks (1D-CNNs). The proposed method has been validated with experimental work not only for single damage but also for multiple damage cases introduced onto ball bearings in laboratory environment. The two 1D-CNNs (one set for the interior bearing ring and another set for the exterior bearing ring) were trained and tested under the same conditions for torque and speed. It is observed that the proposed system showed excellent performance even with the severe additive noise. The proposed method can be implemented in practical use for online defect detection, monitoring, and condition assessment of ball bearings and other rotatory machine elements. © 2022, The Society for Experimental Mechanics, Inc.
Dynamic response levels are critical for tall and slender civil structures. Studying the dynamic behavior of large civil structures with finite element modeling techniques requires detailed and accurate modeling of structural geometry, material properties, member fixities, connection types, and accompanying assumptions. Still, the finite element model results are approximations that could be away from representing the actual structural behavior. Structures are dynamically tested at their operational conditions to validate the finite element model results. This paper presents Operational Modal Analysis (OMA) and finite element model updating of a tall structure located in the West Bay area of Doha (Qatar). The structure is a reinforced concrete building with shear wall cores situated towards the center of the building plan, which was constructed between 2012 and 2016. With 53 stories above the ground and two stories below ground, the 230 m (755 ft) tall building is being used for residential and hotel purposes. For the finite element model updating and calibration tasks presented in this paper, the authors intentionally introduced drastic model changes for the first two model updates so that the results from the first two attempts guide how to proceed with a more reasonable update for the third calibration of the finite element model. While this is a non-standard technique that represents a specific condition where the initial attempts on the finite element model are very crude approximations, it is a systematized demonstration of how to operate when the structural parameters are sparse or uncertain for modeling purposes. While in theory, the finite element model updates can always be fine-tuned in a way to further decrease the error between the measured and predicted OMA results, in this paper, the authors predominantly focused on the presentation of three finite element model updates to demonstrate the way they have improved the modal assurance criteria plots and lowered the average absolute errors by visiting two drastic and then one moderate finite element model updates. The material presented here in this paper is arguably the first published work on large-scale dynamic testing of a civil structure in the State of Qatar. © 2021 Institution of Structural Engineers
This paper presents Operational Modal Analysis (OMA) and Finite Element (FE) model updating of a tall structure. Located in the West Bay area of Doha (Qatar), the structure was constructed between 2012 and 2016. It is a reinforced concrete building with shear wall cores located towards the center of the building plan. With 53 stories above the ground and 2 stories below ground, the 230 m (755 ft) tall building is being used for residential and hotel purposes. The material presented here is arguably the first published work on large-scale dynamic testing of a civil structure in Qatar. The wireless sensors used for testing are state-of-the-art equipment that can capture very low frequencies, something that cannot be accomplished with most of the conventional accelerometers available in the market. © 2022, The Society for Experimental Mechanics, Inc.
Stadiums like those used for sporting or concert events are distinct from other civil engineering structures due to several different characteristics. Some challenges mainly originate from the interaction with the human factor, as stadiums are subjected to both synchronized and random motion of large crowds. The investigations in the literature on this topic clearly state that stadiums designs are in urgent need of more reliable load quantification and modeling strategies, deeper understanding of structural response, generation of simple but efficient human–structure interaction models, and more accurate criteria for vibration acceptability. Although many esthetically pleasing and technologically advanced stadiums have been designed and constructed using structurally innovative methods, recent research on this field still calls for less conservative and more realistic designs. This article aims to highlight the recent advances in this field and to provide a follow-up to the literature review covering until 2008 (Jones et al., 2011a) on vibration serviceability of stadiums structures. The article will also discuss new sensing and monitoring techniques on load-time history measurements and their regeneration, as well as crowd motion, stadium health monitoring, and human comfort analysis. Operational effects of crowds on the dynamic properties are also discussed. The article concludes with a forward look on the recommended work and research for dynamic assessment of stadiums.
Unlike most of civil engineering structures whose static and dynamic responses are estimated accurately through several codes and guidance, stadiums reserve a distinctive place especially when it comes to their dynamic behavior. This difference takes its source from several factors such as influence of crowd size, motion and slenderness of the structure. The most noticeable form of this difference shows itself as excessive vibration levels which is actually a threat to the serviceability of these structures. Eventually, it becomes essential to carefully evaluate several steps of this particular problem starting from correct representation of crowd activity through accurate loadings and human-structure interaction models to arranging acceptable vibration serviceability limits. This publication intends to point out the newly developed techniques and discovered issues on several stages of the problem during the last decade.
Annoying vibrations in grandstand structures have been receiving more attention due to the increasing slenderness of the architectural components and the complexity of the crowd loading for engineers. The vibration serviceability checks under these conditions become a challenge in the design and operation stages. Regarding human comfort, excessive vibrations due to occupant activities may affect comfort and/or cause panic, especially for passive occupants who do not participate in generating excitations. Although durations of excessive vibrations have been considered as one of the most important factors affecting occupant comfort, incorporating the vibration duration in the occupant comfort analysis has not been addressed yet. In addition, the currently available approaches using raw acceleration, weighted RMS acceleration, vibration dose values (VDV), and so on may not always be sufficient for serviceability assessment due to the lack of guided procedure for calculating the integration time and implementing the duration of vibration into the process. Therefore this study proposes a new parameter and framework for assessing human comfort which incorporates the duration of vibration with conventional data processing. The aim is to better examine vibration levels and the corresponding occupant response focusing on grandstand structures. A new parameter, the area of RMS (ARMS), is introduced using the running RMS values of acceleration weighted by the frequency weighting functions. Furthermore, perception ranges for human comfort levels based on the proposed parameter are presented. The experimental study reveals that the proposed framework can successfully address the impact of duration time on determining the levels of vibrations and comfort using the proposed parameter.
House Charlie is an office building located in Växjö, Sweden, with approx. 5,700 m2 area on four floors, fitting 3,700 m2 of office space, and 2,000 m2 of restaurants and conference rooms. The load-bearing structure is a column-beam system made from glued laminated timber (Glulam) with the flooring made from cross-laminated timber (CLT). The house is equipped with a net-work of sensors which were already installed during the construction phase. The design of the network was done in collaboration between the Department of Building Technology from Linnaeus University and SAAB, in close contact Videum and JSB, the owner and constructor, respectively. In the network, two sensor cards collect data from the sensor (displacement, relative humidity, temperature, vibrations, as well as weather station data) which is accessible via a 3G-router from the outside. Except for power supply the network is work-ing independently from the buildings facilities. The building was erected and the network installed during spring 2018, since then the network is providing data. The report describes the measurement network and its sensors as well as their positioning within the building. Additionally the results are presented for the time-span July 2018-December 2019 as well as an interpretation of the first 1.5 years of run-time are given.
Wood and wood-based products interact with the surrounding environment. The interaction with moisture is particularly interesting since it influences the structural properties and may lead to degradation. A structural health monitoring system was established in House Charlie, a four-story office building in Växjö, Sweden, made from timber. It has been running since summer 2018, collecting vibration data and information on temperature and humidity at multiple positions within the façade and the slab. The present work shows that the moisture content within a slab of the building varies throughout the years, attributed to an ongoing dry-out and seasonal changes. Furthermore, the variation is directly coupled with the weather data from a public weather station in Växjö.
Wood is widely used in the construction sector and gaining increased market share. It isinteracting with the surrounding so that its mechanical and geometrical properties (stiffness,strength, swelling, density, …) change with temperature and humidity levels. In a full-scalebuilding, the eigenfrequencies are hence also varying with the climate. In the current paper,results from a preliminary experimental study are presented. A beam made from cross-laminated timber was hanging freely supported inside a climate chamber. Enforced vibrationsfrom a controlled shaker were taken to obtain the eigenfrequencies. With decreasing moisturecontent, the first and third eigenfrequencies were increasing (bending modes) while the secondeigenfrequency was decreasing (torsional mode). A finite element study allowed for checkingwhich parameters is influencing to which degree so that individual changes can be combined.
Korslimmat trä (KL-trä) tillverkas av brädor som limmas ihop korsvis i flera skikt. Skivorna som erhållsanvänds som byggelement, mestadels för väggar och bjälklagselement. Utgångsmaterialet trä tarstår i jämvikt med det omgivande klimatet och kommer därför ändra fuktkvoten. I studien undersöktesen skiva av KL-trä under varierande fuktförhållanden i en klimatkammare hos Linnéuniversitetet.Egenfrekvenser samt fuktkvoten följdes upp och sambandet med klimatet studerades. Det visade sigatt första och tredje uppmätta egenfrekvensen (böjning) visade negativ korrelation med omgivandefukten, den gick upp när fukten minskades (och tvärtom). För andra egenfrekvensen (torsion)däremot visade sig ett mer komplicerat samband. Ett flertal möjliga orsaker presenteras som förklaring. Medverkande organisationer var Linnéuniversitetet som huvudpart samt SödraSkogsägarna och Saab som bidragit som stödfunktionen och bollplank.
Metamaterials (MMs) are composites that are artificially engineered to have unconventional mechanical properties that stem from their microstructural geometry rather than from their chemical composition. Several studies have shown the effectiveness of viscoelastic MMs in vibration attenuation due to their inherent vibration dissipation properties and the Bragg scattering effect. This study presents a multiobjective optimization based on genetic algorithms (GA) that aims to find a viscoelastic MM crystal with the highest vibration attenuation in a chosen low-frequency range. A multiobjective optimization allows considering the attenuation due to the MM inertia versus the Bragg scattering effect resulting from the periodicity of the MM. The investigated parameters that influence wave transmission in a one-dimensional (1D) MM crystal included the lattice constant, the number of cells and the layers' thickness. Experimental testing and finite element analysis were used to support the optimization procedure. An electrodynamic shaker was used to measure the vibration transmission of the three control specimens and the optimal specimen in the frequency range 1-1200 Hz. The test results demonstrated that the optimized specimen provides better vibration attenuation than the control specimens by both having a band-gap starting at a lower frequency and having less transmission at its passband.
In the woodworking industry, detection of annual rings and location of pith in relation to timber board cross sections, and how these properties vary in the longitudinal direction of boards, is relevant for many purposes such as assessment of shape stability and prediction of mechanical properties of timber. The current work aims at developing a fast, accurate and operationally simple deep learning-based algorithm for automatic detection of surface growth rings and pith location along knot-free clear wood sections of Norway spruce boards. First, individual surface growth rings that are visible along the four longitudinal sides of the scanned boards are detected using trained conditional generative adversarial networks (cGANs). Then, pith locations are determined, on the basis of the detected growth rings, by using a trained multilayer perceptron (MLP) artificial neural network. The proposed algorithm was solely based on raw images of board surfaces obtained from optical scanning and applied to a total of 104 Norway spruce boards with nominal dimensions of 45×145×4500mm3. The results show that optical scanners and the proposed automatic method allow for accurate and fast detection of individual surface growth rings and pith location along boards. For boards with the pith located within the cross section, median errors of 1.4 mm and 2.9 mm, in the x- and y-direction, respectively, were obtained. For a sample of boards with the pith located outside the board cross section in most positions along the board, the median discrepancy between automatically estimated and manually determined pith locations was 3.9 mm and 5.4 mm in the x- and y-direction, respectively.
A computer-implemented method for generating a training dataset for training an artificial neural network configured to use images of lateral faces of a timber board to provide information about structure and/or defects, the method including; a log generation step during which a virtual model of a log is generated; a sawing step of the virtual model to obtain one or more virtual timber boards; a pattern step during which a surface pattern is determined as the intersection between the virtual lateral face and the internal structure and/or defects; a rendering step during which a rendered surface image of the lateral face of the virtual timber board is created; and an input data generation step during which the rendered surface images are used to create one or more item of input data; an output data generation step during which an item of output data is generated; and a population step during which a record is added to the training dataset comprising the item of input data, in combination with the item of output data.
A computer-implemented method for estimating a pith location with regard to a timber board, including: receiving a pixelated actual digital image of each lateral face of at least a longitudinal part of the timber board, extending along a longitudinal axis of the timber board; identifying an input portion in said longitudinal part of the timber board, where the input portion is a portion of the timber board which extends along the longitudinal axis; extracting from each pixelated actual digital image of the longitudinal part of the timber board, an input image representing said input portion, so obtaining four input images representing an appearance of the input portion at each lateral face of the timber board; inputting said four input images into the input layer of an artificial neural network and making the artificial neural network operate; and reading, at an output layer of the artificial neural network, output data defining a location of a pith of a log from which the timber board has been obtained, in a plane perpendicular to the longitudinal axis of the timber board at the input portion.
Knowledge of pith location is needed for modelling of sawn timber and for real time assessment of wood material in the wood working industry. However, the methods that are available and implemented in optical scanner today seldom meet customer requirements on accuracy and/or speed. In the present research data of greyscale images of the four longitudinal sides of board and a one-dimensional convolutional neural network were used to determine pith location along Norway spruce timber boards. A novel stochastic model was developed to generate thousands of virtual timber boards, with photo-realistic surfaces and known pith location, by which the network was trained before it was successfully applied to determine pith location along real boards.
Optical scanning and X-ray computed tomography (CT) scanning of sawn timber provide a large numberof data points, on which data-driven numerical models can be based for simulations. These models require informationabout the deviations of the fibre orientations in the vicinity of knots. Optical scanning can be used to measure the in-planefibre orientation on wood surfaces. In CT scans of sawn timber, the fibre orientation around knots can be estimated usinga new fibre reconstruction algorithm based on the density gradient. The goal of this paper is to compare and synchroniseoptical and CT scanning data of sawn timber and then use the combined data set to evaluate fibre orientations derivedfrom both representations. The material comprised sawn timber of Norway spruce, in which alignment holes were drilled.The timber was scanned in an industrial CT scanner and subsequently in an industrial optical scanner where scanning wasrepeated after successive planing of the sawn timber surface. The results show that a projective mapping in combinationwith a spline interpolation are required for synchronisation, and that the in-plane fibre orientations calculated from thedensity gradients are qualitatively similar to the orientations derived from the optical data.
During the last decade, Convolutional Neural Networks (CNNs) have become the de facto standard for various Computer Vision and Machine Learning operations. CNNs are feed-forward Artificial Neural Networks (ANNs) with alternating convolutional and subsampling layers. Deep 2D CNNs with many hidden layers and millions of parameters have the ability to learn complex objects and patterns providing that they can be trained on a massive size visual database with ground-truth labels. With a proper training, this unique ability makes them the primary tool for various engineering applications for 2D signals such as images and video frames. Yet, this may not be a viable option in numerous applications over 1D signals especially when the training data is scarce or application specific. To address this issue, 1D CNNs have recently been proposed and immediately achieved the state-of-the-art performance levels in several applications such as personalized biomedical data classification and early diagnosis, structural health monitoring, anomaly detection and identification in power electronics and electrical motor fault detection. Another major advantage is that a real-time and low-cost hardware implementation is feasible due to the simple and compact configuration of 1D CNNs that perform only 1D convolutions (scalar multiplications and additions). This paper presents a comprehensive review of the general architecture and principals of 1D CNNs along with their major engineering applications, especially focused on the recent progress in this field. Their state-of-the-art performance is highlighted concluding with their unique properties. The benchmark datasets and the principal 1D CNN software used in those applications are also publicly shared in a dedicated website. While there has not been a paper on the review of 1D CNNs and its applications in the literature, this paper fulfills this gap. (C) 2020 The Author(s). Published by Elsevier Ltd.
Continuous long-term monitoring of motor health is crucial for the early detection of abnormalities such as bearing faults (up to 51% of motor failures are attributed to bearing faults). Despite numerous methodologies proposed for bearing fault detection, most of them require normal (healthy) and abnormal (faulty) data for training. Even with the recent deep learning (DL) methodologies trained on the labeled data from the same machine, the classification accuracy significantly deteriorates when one or few conditions are altered, e.g., a different speed or load, or for different fault types/severities with sensors placed in different locations. Furthermore, their performance suffers significantly or may entirely fail when they are tested on another machine with entirely different healthy and faulty signal patterns. To address this need, in this pilot study, we propose a zero -shot bearing fault detection method that can detect any fault on a new (target) machine regardless of the working conditions, sensor parameters, or fault characteristics. To accomplish this objective, a 1D Operational Generative Adversarial Network (Op-GAN) first characterizes the transition between normal and fault vibration signals of (a) source machine(s) under various conditions, sensor parameters, and fault types. Then for a target machine, the potential faulty signals can be generated, and over its actual healthy and synthesized faulty signals, a compact, and lightweight 1D Self-ONN fault detector can then be trained to detect the real faulty condition in real time whenever it occurs. To validate the proposed approach, a new benchmark dataset is created using two different motors working under different conditions and sensor locations. Experimental results demonstrate that this novel approach can accurately detect any bearing fault achieving an average recall rate of around 89% and 95% on two target machines regardless of its type, severity, and location.
1D Convolutional Neural Networks (CNNs) have recently become the state-of-the-art technique for crucial signal processing applications such as patient-specific ECG classification, structural health monitoring, anomaly detection in power electronics circuitry and motor-fault detection. This is an expected outcome as there are numerous advantages of using an adaptive and compact 1D CNN instead of a conventional (2D) deep counterparts. First of all, compact 1D CNNs can be efficiently trained with a limited dataset of 1D signals while the 2D deep CNNs, besides requiring 1D to 2D data transformation, usually need datasets with massive size, e.g., in the "Big Data" scale in order to prevent the well-known "overfitting" problem. 1D CNNs can directly be applied to the raw signal (e.g., current, voltage, vibration, etc.) without requiring any pre- or post-processing such as feature extraction, selection, dimension reduction …