Structural Health Monitoring Methods Research Articles

Development and experimental validation of a prototype system for Machine Learning based SHM in composite aerostructures

The development and experimental validation of a Machine Learning (ML) vibration based Structural Health Monitoring (SHM) prototype system for composite aerostructures operating under uncertainty is presented. Following initial training, it is intended to operate autonomously, in almost real-time, without interrupting normal operation, relying on a limited number of acceleration and strain measurements from properly selected locations on the aerostructure. It is equipped with a multi-level information fusion methodology comprising multiple sensor technologies and diagnostic algorithms for optimized SHM performance in terms of damage detection and characterization. Through its flexible software, the system provides multiple options of ML based SHM methods targeting wide applicability to various types of structures. The system is validated based on numerous experiments with healthy and damaged full-scale composite bonded spars operating under multiple uncertainty factors. Its diagnostic performance indicates that the current progress in hardware and robust vibration-based Machine Learning algorithms may lead to compact, effective, and low cost SHM systems for composite aerostructures under normal operating conditions.

Advanced methods for the detection of measurement outliers in fibre optic strain measurements for non-destructive monitoring of the structural health of composite over-wrapped pressure vessels of type-IV

To analyse the applicability of fibre optic measurements as a measurement method for structural health monitoring, fibre-optic sensors are placed on the vessel surface. The influence of the surface profile of the vessel on the measurement result is analysed by correlating the 3D scanned surface and measurement results. One outcome is that the measurement is substantially influenced by the surface topography. The surface height can be directly correlated to the strain magnitude, which is also verified by analytical calculations. Outliers and invalid values can be traced back to surface-related influences on the sensor fibre. For the outlier detection, a novel method is presented, which is based on quantitative evaluation criteria and inductive reasoning and enables transient cycle- by-cycle outlier detection adapted to the load case. Based on the results an estimate for the expanded measurement uncertainty is given for of the measurement chain consisting of vessel, adhesive and sensor fibre.

Adaptive crack damage identification based on multi-scale sample entropy under variable temperature environment

High-speed trains inevitably experience the impact of complex temperature loads during operation, which can easily result in cracks. Structural health monitoring methods based on Lamb waves are affected by the impact of temperature on their propagation mechanism, leading to difficulties in constructing accurate damage diagnostic models. This paper proposes a crack damage identification method based on adaptive multi-scale sample entropy under variable temperature environment. A sliding window method is proposed to obtain multiple wave packets in a Lamb wave, which avoids the disadvantage of insufficient information of a single wave packet and improves the robustness and reliability of diagnosis. A variance-based multiscale transform method is proposed to process Lamb wave signals, which reduces the sensitivity of Lamb wave signals to temperature compared to the traditional mean-based multiscale transform method. Lamb waves in different sliding windows are transformed at multiple scales to extract damage Multi-scale sample entropy (MSE) at different scales. The Quantum genetic algorithm (QGA) is introduced to optimize sliding windows and MSE, achieving adaptive MSE extraction under variable temperature environments. Based on the advantage of adaptive multiscale entropy, a quantitative crack diagnosis model is established. To verify the effectiveness of the proposed method, crack detection experiments under variable temperatures were conducted. The results show that the proposed adaptive MSE has good resistance to temperature changes and can accurately identify crack length.

A new PZT-enabled device for monitoring prestress loss in post-tensioned prestressed structures

ABSTRACT Anchorage performance monitoring plays an important role in ensuring the safety of prestressed structures. Over the past decades, traditional structural health monitoring (SHM) methods, such as vibration, magnetoelastic, and acoustoelastic methods, have been plagued by low sensitivity and high susceptibility to environmental and temperature interference. Lead-Zirconate-Titanate (PZT)-enabled active sensing method has proven its effectiveness in prestress monitoring. The main contribution of this paper is the development of a new device that utilises the relationship between energy transfer and contact stress at the interface. This device addresses the existing problem of signal saturation and offers higher sensitivity for practical implementation. First, the time reversal (TR) method was used to overcome the problem of low signal-to-noise ratio (SNR) in traditional methods. Based on this, the empirical mode decomposition (EMD) method was employed to process the acquired signals, thereby improving the stability of the data. A quantitative index for monitoring prestress loss was proposed by normalising the peak value of the reconstructed signals. Finally, several repeated experiments were conducted to verify the accuracy of the proposed method.

Detection, Localization, and Quantification of Damage in Structures via Artificial Neural Networks

This paper presents a structural health monitoring method based on artificial neural networks (ANNs) capable of detecting, locating, and quantifying damage in a single stage. The proposed framework employs a supervised neural network model that uses input factors calculated by modal parameters (natural frequencies or mode shapes), and output factors that represent the damage situation of elements or regions in a structural system. Unlike many papers in the literature that test damage detection methods only in numerical examples or simple experimental tests, this work also assesses the presented method in a real structure showing that it has potential for applications in real practical situations. Three different cases are evaluated through the methodology: numerical simulations, an experimental lab structure, and a real bridge. Initially, a cantilever beam and a 10-bar truss were numerically analyzed under ambient vibrations with different damage scenarios and noise levels. Afterward, the method is assessed in an experimental beam structure and in the Z24 bridge benchmark. The numerical simulations showed that the methodology is promising for identifying, locating, and quantifying single and multiple damages in a single stage, even with noise in the acceleration signals and changes in the first vibration mode of 0.015%. In addition, the Z24 bridge study confirmed that the damage detection method can localize damage in real civil structures considering only natural frequencies in the input factors, despite a mean difference of 4.08% between the frequencies in the healthy and damaged conditions.

A multi-layer ML model evolutionary paradigm for high-accuracy individual aircraft SHM

Evolutionary Machine Learning (EML) has emerged as a novel direction for providing robust methodologies for effectively tackling complex problems. However, most current research focuses on multi-objective optimization problems in machine learning (ML). There is limited research on the EML mechanism along the temporal application process, where a fixed machine learning model trained with the prior dataset will inevitably be not representative of the target problem with time. Taking the Structural Health Monitoring (SHM)-ML damage diagnosis model as an example, this paper proposes a novel multi-layer ML model evolutionary paradigm with multi-source data for high-accuracy individual aircraft SHM. The proposed paradigm considers three interconnected layers for the SHM-ML damage diagnosis model, including the dataset to establish the diagnosis model, the model structure and its parameter training, and mapping relationship used in the model to give the damage size. Besides, for a specific model of aircraft, there exist multiple stages from design to service, where multi-source of data is fused for the multi-layer evolution, enabling continuous improvement of damage diagnosis accuracy. Experimental validation is conducted on critical connection structures of aircraft, specifically the attachment lug structure, using the active guided wave SHM method. The results demonstrate the effectiveness of the proposed multi-layer ML model evolutionary paradigm.

Deep Learning Methods for Vibration-Based Structural Health Monitoring: A Review

Deep Learning Methods for Vibration-Based Structural Health Monitoring: A Review

Advancing Bridge Structural Health Monitoring: Insights into Knowledge-Driven and Data-Driven Approaches

Structural health monitoring (SHM) is increasingly being used in the field of bridge engineering, and the technology for monitoring bridges has undergone a radical change. It has evolved from the initial local monitoring and assessment, which relied mainly on manual work, to the current all-round and full-time intelligent assessment provided by intelligent monitoring systems. This paper reviews the development of structural health monitoring technology in the civil engineering field and examines two current AI methods in bridge structural health monitoring, namely knowledge-driven and data-driven approaches. The advantages and disadvantages of these two AI methods are analyzed, and future development trends are also discussed. The overview results reveal that knowledge-driven methods have the advantages of interpretability and stability. However, their current application is limited, and significant technical bottlenecks remain. On the other hand, the data-driven approach demonstrates higher efficiency and accuracy. Nevertheless, it is characterized by instability and insecurity due to its "black-box" nature, which hinders its ability to explain the internal operation mechanism. Given these findings, the hybrid knowledge-data-driven approach emerges as a potential solution. This approach can effectively integrate the advantages of both knowledge-driven and data-driven methods while avoiding their respective disadvantages. Consequently, the hybrid approach proves to be more stable, safe, and efficient in practical applications.

Debonding Detection in Grouted Sleeves Using Axisymmetric Longitudinal Guided Waves.

Grouted sleeves (GSs) are a type of precast joint that can effectively connect steel rebars with excellent performance. However, the grouting debonding problem, which can occur due to the leakage of the glue plug, can seriously affect the properties of GSs. In this paper, a guided-wave-based structural health monitoring (SHM) method is used to detect debonding in GSs. The axisymmetric longitudinal mode is selected as the incident wave since it is sensitive to axial damage. Eight piezoelectrics (PZTs) are then symmetrically installed to actuate signals. The proposed samples are GSs with four different debonding sizes. First, the relationship between the arrival time of the first wave packet and the debonding size is explored through theoretical derivation. The arrival time decreases linearly with an increasing debonding size. A similar trend is observed when the relationship is verified via a numerical simulation and experimental results. This method will provide a reference for detecting debonding in similar GS multilayer structures.

Hierarchical verification and validation in a forward model-driven structural health monitoring strategy

This paper presents a demonstrative application of a forward model-driven approach to structural health monitoring (SHM), incorporating hierarchical validation methods. A key tenet of the approach is that an SHM system can be constructed that is capable of diagnosing damage at the full system level, without full system damage-state data having been used in its development; achieving this would be highly impactful as the system-level damage state data is generally not feasible to acquire (previous SHM methods such as data-driven SHM have been hindered by their dependence on these data). This is achieved by carrying out validation activities on the damage model at the subassembly level of the structure. The particular focus of the present paper is on damage detection and assessment, although the approach offers a natural basis for extension to other damage identification activities such as damage location and prognosis. The present study focuses on two of the key elements of the model-driven approach: validation of the predictive substructure models and their application in the assembled model. The ideas discussed are demonstrated in a case study based on a laboratory-scale truss bridge structure.

Research on structural health monitoring methods for bridge structures

With the development of science and technology, China is building more and more Bridges. The inspection and maintenance of bridge during operation plays an important role in the safety of bridge. Firstly, the research background and significance of bridge structure health inspection are introduced, and the research status and progress of health inspection are analyzed. Then, several commonly used monitoring methods are listed. Based on the comprehensive study of the bridge health monitoring system, the main conclusions are as follows: the bridge health monitoring system is realized through the coordination of several subsystems, including: bridge working environment monitoring, bridge structure overall performance monitoring and bridge structure local performance monitoring. The bridge structural health monitoring system is mainly used in bridge damage identification and bridge safety warning. The bridge structure health monitoring system conforms to the development trend of high technology and will become an indispensable part of the bridge operation in the future.

An enhanced topological analysis for Lamb waves based SHM methods

Topological data analysis (TDA) is a powerful and promising tool for data analysis, but yet not exploited enough. It is a multidimensional method which can extract the topological features contained in a given dataset. An original TDA-based method allowing to monitor the health of structures when equipped with piezoelectric transducers (PZTs) is introduced here. Using a Lamb wave based Structural Health Monitoring (SHM) approach, it is shown that with specific pre-processing of the measured time-series data, the TDA (persistent homology) for damage detection and classification can be greatly improved. The TDA tool is first applied directly in a traditional manner in order to use homology classes to assess damage. After that, another method based on slicing the temporal data is developed to improve the persistence homology perception and to leverage topological descriptors to discriminate different damages. The dataset used to apply both methods comes from experimental campaigns performed on aeronautical composite plates with embedded PZTs where different damage types have been investigated such as delamination, impacts and stiffness reduction. The proposed approach enables to consider a priori physical information and provides a better way to classify damages than the traditional TDA approach. In summary, this article demonstrates that manipulating the topological the features of PZTs time-series signals using TDA provides an efficient mean to separate and classify the damage natures and opens the way for further developments on the use of TDA in SHM.

Baseline-Free Detection Method for Change of Lateral Stiffness of High-Rise Building Based on Statistical Moment Curvature

In recent times, there has been a notable increase in the quantity of high-rise buildings, attributed to the swift advancements in both economic growth and construction technology. Assessing the structural integrity of high-rise buildings is important to ensure their safe operation. However, existing structural health monitoring methods typically require a baseline, involving either the measured dynamic and static responses from an intact structure or the finite element (FE) model corresponding to an undamaged state. These prerequisites are often challenging to acquire in practical scenarios. This study introduces a novel baseline-free method for detecting reduction in the lateral stiffness of high-rise buildings. The method is based on the statistical moment curvature (SMC) concept, determined through applying central difference to the second-order statistical moment of displacement. Initially, theoretical formulas were derived to demonstrate the viability of utilizing SMC for identifying reduction in the lateral stiffness of high-rise buildings. Subsequently, a FE model of a representative high-rise building was constructed to validate the proposed approach and assess its sensitivity, where different structural types and noise levels were considered. Lastly, a field test was conducted on a 33-story shear wall structure to provide additional validation for the proposed method. The results confirmed its effectiveness in accurately detecting reduction in the lateral stiffness of high-rise building. It offers two primary benefits: firstly, it obviates the need for a baseline, rendering it more convenient and applicable in real-world scenarios; secondly, its heightened sensitivity to sudden drops in lateral stiffness allows for early-stage detection of structural damage.

A method to maximise the information obtained from low signal-to-noise acceleration data by optimising SSI-COV input parameters

Structural Health Monitoring (SHM) has mainly been undertaken on larger bridges and on a case-by-case basis. This is due to a range of factors, such as the high installation costs and the effort required to install and commission the monitoring systems. One way in which SHM systems can become more feasible for widespread adoption at a network level is to reduce the number and cost of sensors used. However, this comes with a trade-off as low-cost sensors will typically have a worse signal-to-noise ratio and the reduced number of sensors requires careful placement to maximise the amount of information acquired. One of the simplest/cheapest methods of bridge SHM is long-term tracking of the bridge frequency to identify a change in stiffness. Using data collected from five in-service bridges, this research shows that the user-defined SSI-COV input parameters can significantly impact the quality of extracted natural frequencies. Consequently, a novel method is developed to aid in choosing the inputs used in the SSI-COV method. The method developed also showed that the determined inputs resulted in the extraction of accurate natural frequencies with minimal apparent outliers on all tested bridges. The developed method allows the extraction of quality natural frequencies from low signal-to-noise acceleration data which are vital when undertaking frequency-based SHM.

Structural Condition Monitoring Using Deep Learning on a Metallic Part Fabricated by Additive Manufacturing

Additive manufacturing (AM) was originally developed to manufacture polymer prototypes. Today, it has been used for the manufacturing of many critical machine components. Most of the structural health monitoring (SHM) methods were developed for monitoring the condition of large and thin plates on airplane fuselages. Additively manufactured parts are generally small, thick, and have complex geometries. SHM methods have been improved to sense load, detect defects, and identify loose bolts with the help of a permanently installed sensor. In this study, the adaptability of SHM methods was researched with additively manufactured metal parts with complex geometry. Magnets were used to apply pressure to 9 different locations on the surface of a stainless steel additively manufactured thick plate with deep groves. SHM was used to estimate the magnets’ location. Many SHM (Lamb wave) methods cannot work on smaller parts since their dimensions are shorter or very close to the wavelength of the created oscillations. Surface response to excitation (SuRE) method which has similar characteristics to electromechanical impedance methods was used for data collection. To obtain descriptive features of the time domain data, fast Fourier transformation (FFT), short-time Fourier transformation (STFT), continuous wavelet transformation (CWT), and synchrosqueezing transform (SST) were applied to CWT. 1D and 2D convolutional neural networks (CNN) were used to classify the cases. When CNN was optimized for the analysis of our data, 100% location estimation accuracy was obtained by using 50% of 320 scalograms for training. The scalograms were obtained by enhancing the CWT results with SST. STFT-CNN combination was the second best. It obtained 95% accuracy with the same number of spectrograms and training allocation.

System identification and finite element model updating of a 6 MW offshore wind turbine using vibrational response measurements

Offshore wind energy is playing an increasingly vital role in the clean energy transition around the world, and improved reliability of wind turbine structures is necessary for the long-term success and efficiency of renewable energy. Increased reliability would reduce costs associated with maintenance due to breakages and in turn reduce the levelized cost of energy for offshore wind energy sources. Structural health monitoring methods can be used to predict breakages and extend lifetimes by continuously monitoring instrumented structures. This paper presents system identification and model updating of a 6 MW offshore wind turbine using vibration measurements under varying operational conditions. The turbine is monopile-supported and instrumented with strain gauges and accelerometers at several elevations along the tower and monopile. Effective stiffness of soil springs in the model are updated to match modal-predicted natural frequencies and mode shapes of the first two modes with those identified from measurements at different operating conditions. A deterministic and probabilistic (Bayesian) approach to model updating are compared. The sensitivity of identified modal parameters and the updated model parameters are investigated with respect to operational and environmental conditions such as wind speed. Results show that deterministic model updating can match modal parameters with high accuracy across datasets and environmental conditions. Bayesian model updating results successfully estimate the posterior distribution of updating model parameters with an increasing degree of certainty as more data is used.

Application of Vehicle-Based Indirect Structural Health Monitoring Method to Railway Bridges—Simulation and In Situ Test

In recent years, the vehicle-based indirect Structural Health Monitoring (iSHM) method has been increasingly used to identify the dynamic characteristics of railway bridges during train crossings, and it has been shown that this method has several advantages compared to traditional SHM methods. A major advantage is that sensors are just mounted on the vehicle, and no sensors or data acquisition systems need to be installed on the railway bridge. In this paper, the application of the vehicle-based iSHM method is demonstrated numerically and experimentally for determining the natural frequencies of railway steel bridges during train crossing. The coupled linear equations of motion of the train-bridge multi-body system are derived, and train crossing simulations are conducted numerically, considering different train speeds. Three different railway bridges are considered, and the train-induced vibration responses are calculated for both the train multi-body system and the railway bridge models. Different representative evaluation points are chosen for the wheelsets, bogies, and car bodies of the considered vehicle. To calibrate the numerical model, the resonance frequencies of an existing single-span steel bridge are measured in situ by the application of forced vibration tests. Besides the executed in situ measurements of the bridge, the considered crossing vehicle is also instrumented with several accelerometers at the wheelsets, bogies, and car bodies, and the vibration responses of both the bridge and the crossing vehicle are measured simultaneously during the duration of several train crossings with different train speeds. The recorded vibration responses are analyzed in the frequency domain and compared with numerical simulation results. It is shown that the first bending frequency of the considered railway bridge can be clearly identified from the computed frequency response spectra and that the vehicle-based iSHM method provides a promising tool for identifying the dynamic characteristics of railway bridges.

Assessing Weak Adhesion in Single Lap Joints Using Lamb Waves and Machine Learning Methods for Structural Health Monitoring

The use of adhesive joints has become increasingly popular in various industries due to their many benefits, such as low weight and good mechanical performance. However, adhesive joints can suffer from defects, one of them being weak adhesion. This defect poses a significant risk to structural integrity and can lead to premature failure, but is hard to detect using existing nondestructive testing methods. Therefore, there is a need for an effective technique that can detect weak adhesion in single-lap joints (SLJ) to prevent failure and assist in maintenance, namely in the framework of structural health monitoring. This paper presents a novel approach utilizing machine learning and Lamb Waves (LW) to determine the level of weak adhesion. Firstly, a numerical model of SLJs with different levels of weak adhesion is created and an original approach is proposed for its validation with data from real samples so that reliable LW data can further be easily generated to train and test any other data-driven algorithm for tackling damage. Secondly, a damage detection method is proposed, based on artificial neural networks and fed with simulated data, to determine the level of damage in SLJs, independent of their location. The results show that the simulation model can be validated with a small set of experimental data, being capable of replicating real damage in SLJs. Additionally, the use of simulated data in the training algorithm can increase the accuracy of the simulation model up to 26% when compared to only considering experimental data. The adopted artificial neural network for detecting weak adhesion emerges as a promising approach, yielding a precision of over 95%. Thus, machine learning and LW data can be used to improve the reliability and accuracy of adhesive bonding quality control, as well function as a technique for structural health monitoring, which can enhance the safety and durability of bonded structures.

A Strain Transfer Model for Detection of Pitting Corrosion and Loading Force of Steel Rebar with Distributed Fiber Optic Sensor.

Steel rebar corrosion is one of the predominant factors influencing the durability of marine and offshore reinforced concrete structures, resulting in economic loss and the potential threat to human safety. Distributed fiber optic sensors (DFOSs) have gradually become an effective method for structural health monitoring over the past two decades. In this work, a strain transfer model is developed between a steel rebar and a DFOS, considering pitting-corrosion-induced strain variation in the steel rebar. The Gaussian function is first adopted to describe the strain distribution near the corrosion pit of the steel rebar and then is substituted into the governing equation of the strain transfer model, and the strain distribution in the DFOS is analytically obtained. Tensile tests are also conducted on steel rebars with artificially simulated corrosion pits, which are used to validate the developed model. The results show that the Gaussian function can be used to describe the strain variation near a corrosion pit with a depth less than 50% of the steel rebar diameter, and the strain distribution in the DFOS analytically determined based on the developed strain transfer model agrees well with the tensile test results. The corrosion pit depth and loading force in the steel rebars estimated based on the proposed model agree well with the actual values, and therefore, the developed strain transfer model is effective in detecting pitting corrosion and loading force in steel rebars.

Monopile Foundation Stiffness Estimation of an Instrumented Offshore Wind Turbine through Model Updating

Rapid development of offshore wind foundation models has resulted in a large number of built structures with generally underestimated foundation stiffness properties and a need to update and validate both the individual structural models and the underlying foundation design frameworks. This paper outlines a structural health monitoring approach, based on the combination of output only structural health monitoring methods and model updating, to estimate foundation stiffness parameters using field monitored data. Field monitoring data from an offshore wind turbine under idling conditions, over a large monitoring period, are presented and operational modal analysis is applied to estimate the modal parameters. Those are compared to modal properties predicted by finite element models, employing either old (API/DNVGL) or new (PISA) foundation design properties, which are calibrated using geotechnical site investigation data. A new approach to interpret seabed level statically equivalent foundation stiffness, in terms of effective lateral and rotational stiffness against load eccentricity, is presented. Seabed level statically equivalent foundation properties are updated by comparison against the observed modal behaviour and the optimised foundation parameters are presented, demonstrating a close match to the predictions of the PISA method.