Making Structural Condition Diagnostics Robust to Environmental Variability

Foreword

As newer and more reliable ways of construction were developed, civilization began to spread out further and retain functional infrastructure for longer periods of time.[...]

PurposeOne of the paramount challenges in the field of structural health monitoring (SHM) is the precise assessment of structural damage. Recently, many methods have been proposed to diagnose damage in structures using time series data. Nevertheless, these techniques are restricted in their ability to capture the bidirectional temporal and spatial relationships among the data. To remedy these shortcomings, this study proposes an advanced deep learning (DL) technique that integrates one-dimensional convolutional neural networks (1DCNN) with bidirectional long short-term memory (BiLSTM) networks, termed BiLSTM-1DCNN.Design/methodology/approachThis study proposes an advanced deep learning (DL) technique that integrates one-dimensional convolutional neural networks (1DCNN) with bidirectional long short-term memory (BiLSTM) networks, termed BiLSTM-1DCNN. The novelty of BiLSTM-1DCNN comes in its effective fusion of 1DCNN’s local feature extraction and BiLSTM’s capacity to capture long-term information from both directions (forward and backward).FindingsValidation using a dataset from the Nam O steel truss bridge demonstrates that the proposed method surpasses conventional models, including 1DCNN (87.9%), LSTM (72.5%), BiLSTM (76.1%) and LSTM-1DCNN (91.1%), achieving an accuracy rate of 96.4%. This approach provides a significant advancement in SHM, paving the way for more accurate and reliable damage detection.Research limitations/implicationsThe study’s limitations include potential overfitting due to the model’s complexity and the high computational resources required. The BiLSTM-1DCNN model’s performance is also influenced by the quality and quantity of the training data as well as the specific characteristics of the dataset. Additionally, while the model performs well in detecting damage in the Nam O steel truss bridge, its generalizability to other types of structures and environmental conditions remains untested.Practical implicationsThe BiLSTM-1DCNN model offers significant practical benefits for structural health monitoring by enhancing damage detection accuracy. Its ability to effectively capture both local features and long-term temporal dependencies makes it a valuable tool for real-time monitoring and early detection of structural issues. This can lead to more timely maintenance actions, potentially reducing repair costs and improving safety. The model’s high accuracy in detecting damage can also support infrastructure management decisions, contributing to better resource allocation and enhanced overall safety of critical structures.Social implicationsThe BiLSTM-1DCNN model’s advanced damage detection capabilities have positive social implications, such as improving the safety and reliability of public infrastructure like bridges. Enhanced structural health monitoring can reduce the risk of catastrophic failures, protecting lives and reducing accidents. By providing more accurate and timely assessments, the model can also lead to more efficient use of maintenance and repairs, ensuring that resources are allocated effectively.Originality/valueThe originality of the BiLSTM-1DCNN model lies in its innovative integration of one-dimensional convolutional neural networks (1DCNN) with bidirectional long short-term memory (BiLSTM) networks. This unique combination leverages 1DCNN’s strength in local feature extraction and BiLSTM’s capacity to capture temporal dependencies in both directions, addressing the limitations of existing methods. The model’s high accuracy in detecting structural damage, as demonstrated by its performance on the Nam O steel truss bridge dataset, highlights its value in advancing structural health monitoring technologies and offers a significant improvement over traditional approaches.

Despite various successful applications of Lead Zirconate Titanate (PZT) transducers in the electromechanical impedance (EMI) technique for structural health monitoring (SHM), fundamental research work on the sensitivity of the PZT transducers for damage detection is still in need. Conventionally, the root mean square deviation (RMSD) is utilized to statistically analyze the differences between the pre-damage and the post-damage electromechanical admittance signatures of a PZT transducer. Recently, another damage index method based on the structural impedance has been derived and successfully used for structural damage diagnosis. This paper presents a comparison of the sensitivity of these two methods to the damages in a concrete structure. An experimental test is carried out on a two-storey concrete frame subjected to base vibrations that simulate the earthquake and underground explosions. A number of PZT transducers are regularly arrayed and bonded to the concrete frame structure. After each phase of vibration loading, the PZT transducers are scanned to acquire the raw signatures of the PZT admittance. Subsequently, the damage indices based on both the raw signatures and the structural impedance extracted from the raw signatures are calculated. Results show that the PZT transducers with different distances away from the damage have different sensitivity to the structural damage. The relation between the damage index and the distance of the PZT transducer away from the damage is studied. Consequently, the sensitivity of the PZT transducers is discussed, and their sensing region in concrete is derived.

Quantification of precursor of damage initiation in composite materials is extremely challenging using an on-board Structural health Monitoring (SHM) system. Here we present the recent advancement in online ultrasonic sensing methodology for precursor damage quantification. Present SHM techniques are incapable of detecting the damage stage at sub wave length scale. In this paper, first we have studied a simple but novel off-board approach for precursor damage state quantification using high frequency ultrasonic image correlation technique. Further proper sensing methodology is established from the information extracted from proposed off-board technique. To correlate the on-board SHM technique with the off-board understanding, high frequency piezoelectric sensors were mounted on the specimen and lamb wave interaction with stiffness reduction is studies to excerpt information of damage incubation. Here, we investigated damage state of Carbonfiber- reinforced polymer (CFRP) composite. Precursor damage states are generally in the form of matrix cracking, deboning and fiber breakage in composites. Here in this work we performed tensile-tensile fatigue testing on ASTM standard specimen at an interval of 10000 cycles until cracks / delamination are developed. At each interval of loading, volumetric ultrasonic scans are performed on the gage area of the specimen using Scanning Acoustic Microscope (SAM). The stiffness degradation is calculated at each pixel points of the selected gage area using nonlocal elastoplasticity theory. We used our novel observable parameter called ‘’Damage Entropy” (DE) which is a measure of material damage, are calculated, and plotted with fatigue loading cycle until visible damage incubates. Damage incubation in the specimen was clearly observed from the Z-scan images from SAM. Advance signal processing technique and sensor selection is used for this process. This study has the potential to correlate the DE from SAM and Damage index (DI) from online Sensor Data for online precursor damage quantification. doi: 10.12783/SHM2015/322

Advancements in fiber-reinforced polymer composite materials damage detection methods: Towards achieving energy-efficient SHM systems

This research applies a bibliometric analysis to the structural health monitoring (SHM) research field over a 15-year period, from 2003 to 2017. The goal of this research is to provide a broader analysis of the field as a whole through a data-driven approach with a bibliometric analysis of SHM literature. The aim of this work is to understand relationships between research themes and communities within the SHM field, and how this landscape has evolved over time. By employing graph theory and social network analysis methods, this study uses a data-driven analysis using a co-word and co-venue analysis based on publication data from over 21,000 conference and journal papers in SHM. Strategic diagrams and network analysis were used to map and visualize the academic landscape of SHM. Additionally, a co-venue analysis was performed to determine the conference and journal research communities within the field at each time segment. The analysis was conducted for three 5-year time periods providing insight on how the communities within the SHM research field have evolved.

Strain is an important parameter reflecting structural state, which is particularly important in the field of structural health monitoring (SHM). It can be used to evaluate the mechanical properties, failure behavior, crack development and residual stress of structural members and materials. It is particularly important in the field of structural health monitoring (SHM). In this paper, a new type of large gauge strain sensor is proposed. The proposed sensor is based on micro vision and uses a camera to capture small displacement. The packaging structure of strain sensor with sensing gauge length of about 50 cm is designed. In order to realize the monitoring in the field environment, the microscopic images in the experiment are obtained by webcam, which has broad application prospects. In the field of view of the camera, the maximum distance that the probe in the sensor can move is certain, but when the sensing gauge distance increases, the accuracy of strain measurement will be improved. The strain data obtained by the sensor is compared with the data obtained by FBG sensors to verify the measurement accuracy. It is found that the measurement accuracy is suitable for SHM of infrastructure.

Structural Health Monitoring (SHM) techniques based on high frequency wave propagation (Acousto-Ultrasonics) are of interest due to their sensitivity to incipient damage in structures. Techniques that use small, low profile transducers allow for in-situ sensing and actuation and are particularly useful for integration on aerospace structures. Additionally, signal processing and interpretation of the data becomes important in the diagnosis process. Commonly, the development of an integrated hardware and software system based on Acousto-Ultrasonics is performed on simple structures such as flat metallic panels. In this study, the curved metallic and composite panels are examined as they provide complexity in both geometry and material used to represent helicopter tail boom structures. Damage at the edges of the curved panel is investigated as these locations typically experience damage related to fatigue. Piezoelectric transducers are mounted to the structures and used to excitation and sensing of the wave modes. The data was acquired and damage indices are used to quantify the effect of two types of damage: bolt torque loss, crack propagation between two bolts.

Unsupervised structural damage assessment method based on response correlations

In the long-term service period, the bridge health monitoring system produced a huge amount of monitoring stress data; proper handling of these data is one of the main difficulties in the field of structural health monitoring, especially to predict the structural stress based on the monitored data. The objectives of this article are to present: (1) a nonlinear dynamic model, (2) a nonlinear mixed Gaussian particle filtering algorithm for predicting the monitored data based on the dynamic model, and (3) an approach combining nonlinear mixed Gaussian particle filtering algorithm with structural health monitoring data to predict the structural stress under uncertainty in real time. And an actual example is provided to illustrate the application and feasibility of the proposed models and methods.

The application of MEMS and nanotechnology (MNT) to the field of structural health monitoring (SHM) is a fairly recent development. The recent change in this focus for MNT has been driven by the need to expand the applications for much of the technologies that were developed in the late 1990s. In addition, many companies desire to expand beyond their target high volume market segments of automotive, wireless communications, and computer peripherals, since these market segments were not as lucrative as first predicted. Most of the aerospace structural health monitoring developmental activity has been sponsored by agencies of the U.S. Government, which serves to pace the examination of these newer technologies to some degree. With that said, efforts are underway by companies such as Acellent Technologies and Blue Road Research to explore various MNT structural health monitoring approaches. The MNT under test include embedded piezoelectric sensors, MEMS accelerometers, time domain region sensors, and topical and embedded single and multi-axis fiber optic Bragg grating sensors. The promise of MNT for the SHM market segment is very enticing. The many wireless communication developments and miniaturization developments of the past five years is very attractive to the SHM community, especially those that are able to reduce the cost and complexity of integration. The main challenge for the community is one of selective integration. That is, certain pieces may be appropriate for SHM systems and certain pieces may not be. The better companies will chose wisely and put forth an approach that can be seamlessly integrated into the larger structure. For over a decade, Blue Road Research has been developing technologies aimed at structural health monitoring of both composite and non-composite parts, through the use of single and multiaxis fiber optic Bragg grating sensors. These sensors are 80 to 120 microns in diameter making them smaller than the diameter of a human hair. Multiaxis fiber optic sensors are able to measure pressure, temperature, axial and transverse strain, chemical properties, corrosion, as well as transverse strain gradients. This technology is easily embedded in between the various layers of the composite structure, during manufacture, without compromising the structural integrity, in order to verify manufacturing parameters during the cure cycle and well as monitor the on-going condition of the composite structure throughout its life time. This paper reviews some of the technical work that has been accomplished during the past two years; specifically the embedding of fiber optic sensors into various composite structures in order to be able to conduct in situ non-destructive evaluation of the curing process and the service life of the component. The fiber optic technology has been developed to the point that it is at a TRL of 6.

With the continuous advancement of infrastructure development, ensuring the health and safety of existing structures has become increasingly critical. Structural Health Monitoring (SHM) plays a vital role in assessing the integrity and longevity of buildings, bridges, and other infrastructures. However, traditional monitoring methods often face challenges related to efficiency, accuracy, and cost-effectiveness. The rapid development of Artificial Intelligence (AI) has opened new opportunities for SHM, offering intelligent, automated, and highly precise monitoring solutions. By integrating AI with SHM, it is possible to enhance detection accuracy, improve predictive maintenance, and significantly reduce operational costs. This paper aims to provide a comprehensive summary of AI technologies that have been successfully applied in the field of SHM. By reviewing current advancements, we seek to contribute to the ongoing development of intelligent monitoring systems and offer new perspectives on the future integration of AI in SHM.

Health monitoring has, over the last 2-3 decades, become a topic of significant interest within the structural engineering research community, but also in the broader areas of civil and mechanical engineering. Whereas the merits of Structural Health Monitoring (SHM) are generally appreciated in qualitative terms there has, as of yet, not been reported much fundamental research on the quantification of the benefit of SHM. The present paper addresses the quantification of SHM with basis in the Bayesian pre-posterior decision analysis. The starting points are historical and recent developments in the fields of SHM and the quantification of value of information as well as the identification of typical situations in structural engineering in which SHM has the potential to provide value beyond its costs. Subsequently, the theoretical framework which allows for the quantification of the value of information collected through SHM systems is developed and elaborated. It is shown how the value of information can be quantified to support the assessment and optimization of decisions on whether and how to implement SHM. To illustrate the use of the developed theoretical framework for evaluating the benefit of SHM an example is provided. The example addresses the life-cycle benefit maximization for offshore jacket structures subject to fatigue crack growth utilizing monitoring of near field fatigue stresses as a means of optimizing risk based inspection and maintenance strategies.

Detection of Subtle Damage in Structures through Smart Signal Reconstruction