Structural Health Monitoring Methodology Research Articles

Negative selection algorithm based methodology for online structural health monitoring

Even small damages in engineered systems can endanger the users or lead to significant economic costs. The detection of damage existence is only the first task within the Damage Identification strategy, nonetheless, it is still very complex, especially for Civil Engineering products. The present study is devoted to the analysis of a global methodology for damage detection based on a recently developed version of the Negative Selection Algorithm (NSA). Given any new monitored data measured from the system, the methodology addresses the one-class classification problem by labelling it as normal or abnormal, thus potentially damaged. The strategy is independent of the engineered system under investigation and of the damage-sensitive features used to assess the state, as long as only two features per time are analysed by the classifier. The methodology includes a set of features meant to tackle typical shortcomings and open issues which might prevent an effective application of NSA to damage detection. Some of them are shared with other soft computing techniques for classifications. Finally, the methodology is applied to the data collected from a scaled masonry arch built and tested at the University of Minho (Portugal). The datasets are used to validate the algorithm features, to compare the NSA with other well-known classification algorithms and to simulate a real monitoring to assess the readiness of the algorithm response.

Damage detection & localization on composite patch repair under different environmental effects

This paper presents a structural health monitoring (SHM) methodology for detecting damage in a composite bonded repair. The application of guided wave based techniques in a step-sanded bonded repair under operational and environmental load is thoroughly investigated. A two step damage detection and localization algorithm is presented, were in the first level the path damage indices (PDIs) for each transducer pair is calculated. The PDIs are then compared to a set threshold (based on the environmental and operational conditions) to increase the reliability of damage detection while reducing false alarm. In addition, a self-diagnosis approach based on electro-mechanical impedance (EMI) measure is proposed to identify the faulty sensors prior to the diagnosis. Once the transducer pairs with possible damage in their path has been selected, the second level of the proposed methodology is damage localization. To address the challenge of edge reflection, complex geometrical shape and layup of the repair patch which introduced anisotropy to the wave propagation, a novel damage detection based on probability imaging technique is proposed. The methodology is developed based on assigning probabilities of damage to the Minimal Intersection Score (MIS) to reduce the path saturation related to each path having the same probability of damage being located anywhere along it. The proposed method, uses a smart sub-division technique based on Voronoi Tessellation which is adaptable to any shape (circular, rectangular, elliptical). The reliability of the proposed method is then demonstrated with experimental results on a composite step-sanded repair subjected to impact damage under vibration and temperature variations, and the choice of input parameters such as wave form and excitation frequency on the probability of detecting damage is demonstrated.

Machine learning paradigm for structural health monitoring

Structural health diagnosis and prognosis is the goal of structural health monitoring. Vibration-based structural health monitoring methodology has been extensively investigated. However, the conventional vibration–based methods find it difficult to detect damages of actual structures because of a high incompleteness in the monitoring information (the number of sensors is much fewer with respect to the number of degrees of freedom of a structure), intense uncertainties in the structural conditions and monitoring systems, and coupled effects of damage and environmental actions on modal parameters. It is a truth that the performance and conditions of a structure must be embedded in the monitoring data (vehicles, wind, etc.; acceleration, displacement, cable force, strain, images, videos, etc.). Therefore, there is a need to develop completely novel structural health diagnosis and prognosis methodology based on the various monitoring data. Machine learning provides the advanced mathematical frameworks and algorithms that can help discover and model the performance and conditions of a structure through deep mining of monitoring data. Thus, machine learning takes an opportunity to establish novel machine learning paradigm for structural health diagnosis and prognosis theory termed the machine learning paradigm for structural health monitoring. This article sheds light on principles for machine learning paradigm for structural health monitoring with some examples and reviews the existing challenges and open questions in this field.

Electromechanical impedance measurements for bone health monitoring through teeth used as probes of a Piezo-device

Bone is a dynamic biological tissue that acts as the primary rigid support of the body. Several systemic factors are responsible for pathologies that negatively affect its structural attributes. Although the bone is in continuous renewal by osteogenesis, metabolic diseases are the most common affectations that alter its natural equilibrium. Different techniques based on ionizing radiation are used for the bone diagnosis restrictively. However, if these are not used adequately, the application could present risks for human health. In this paper, it is proposed and explored a new technique to apply an early-stage diagnosis of bone variations. The technique evaluates bone structural conditions from the teeth (used as probes) by applying a structural health monitoring (SHM) methodology. An experimental procedure is described to identify the stiffness variations produced by mechanical drillings done in prepared bone samples. The identification is carried out applying the electromechanical impedance technique (EMI) through a piezo-actuated device in the frequency spectrum 5–20kHz. Three bone samples with incorporated teeth (three teeth, two teeth, and one tooth) were prepared to emulate a mandibular portion of alveolar bone-PDL (periodontal ligament)-tooth system. Piezo-device was attached to the crown of the tooth with an orthodontic bracket allowing the teeth to act as probes. The electrical resistance measurements were computed with an electrical decoupling approach that improved the detection of the drillings; it was due to the increment of the sensitivity of the signals. The results showed that the bone mass reduction is correlated with statistical indices obtained in specific frequency intervals of the electrical resistance. This work suggests the possibility of a future application addressed to a bone diagnosis in a non-invasive way.

Unsupervised damage clustering in complex aeronautical composite structures monitored by Lamb waves: An inductive approach

Structural Health Monitoring (SHM), i.e. the action of monitoring structures in real-time and in an automated manner, is a major challenge in several industrial fields such as aeronautic. SHM is by nature a very high dimensional data-driven problem that possesses several specificities when addressed as a machine learning problem. First of all data in damaged cases are rare and very costly as the generation of damaged data is not always possible and simulations are not reliable especially when dealing with complex structures. SHM is thus by nature an unsupervised problem. Furthermore, any incoming sample should be instantaneously clustered and handcrafted damage indexes are commonly used as a first dimension reduction step due to large datasets to be processed. As a consequence, unsupervised dimensionality reduction (DR) techniques that project very high dimensional data into a two or three-dimensional space (such as t-SNE or UMAP) are very appealing in such a context. However, these methods suffer from one major drawback which is that they are unable to cluster any unknown incoming sample. To solve this we propose to add inductive abilities to these well know methods by associating their projection bases with Deep Neural Networks (DNNs). The resulting DNNs are then able to cluster any incoming unknown samples. Based on those tools, a SHM methodology allowing for unsupervised damage clustering with dimensionality reduction is presented here. To demonstrate the effectiveness of the method, results of damage classification on large experimental data sets coming from complex aeronautical composite structures monitored through Lamb waves are shown. Furthermore, several DR techniques have been benchmarked and recommendations are derived. It is demonstrated that the use of raw Lamb wave signals instead of the associated damage indexes is more effective. This non-intuitive result helps to reduce the gap between laboratory research and the actual start-up of SHM activities in industrial applications.

Structural damage detection and localization using decision tree ensemble and vibration data

AbstractThis paper explores the capabilities of decision tree ensembles (DTEs) for detecting and localizing damage in structural health monitoring (SHM). Unlike research on many other learning models, the goal of this study is to identify damage with a localization accuracy down to the single structural element, rather than limiting the evaluation to the story scale. The SHM methodology herein discussed, denoted as D2‐DTE, is based on decision trees ensemble and belongs to the class of vibration‐based approaches, being the health assessment of the structure obtained by analyzing dynamic properties of the structural system, namely, mode shapes and natural frequencies. The proposed damage detection method is validated for three different test cases, including both numerical simulations and experimentally recorded data, which consider a wide array of damage configurations, including single and multiple damages; different damage types and severities; and the presence of random noise levels associated with dynamic properties acquisition. The performances of the D2‐DTE are evaluated in terms of accuracy, confidence of probabilistic predictions, and measurements of physical distances in localization errors. Additionally, two of the investigated test cases are based on available benchmarks, thus allowing a direct comparison with a state‐of‐the‐art methodology. This comparative analysis evidences competitive performances of the DTE learning method.

Towards the generation of controlled one-inch impact damage in thick CFRP composites for SHM and NDE validation

In this paper, an attempt has been made to conduct controlled low velocity impact experiments in quasi-isotropic carbon fiber reinforced polymer (CFRP) composites of increasing thicknesses to generate barely visible impact damage (BVID) that creates 1″ impact damage diameter. Such impacted CFRP coupons with controlled damage diameter would serve in the future for validation of structural health monitoring (SHM) and nondestructive evaluation (NDE) damage detection methodologies. Various 2-mm, 4-mm and 6-mm thick quasi-isotropic CFRP composite plates having similar stacking sequence were manufactured in a compression molding machine. These plates were cut into numerous 6″ × 4″ coupons to be impacted in accordance with ASTM D7136 standard on a Dynatup impact testing machine. The goal of the experiments was to identify the combination of impactor mass, energy and momentum at which approximately 1″ impact damage diameter could be produced for test coupons of increasing thicknesses. Interesting observations were made with respect to the size and shape of the impact damage as the thickness of the composite coupons was increased from 2-mm to 6-mm. Modified experiments were also conducted on 6-mm coupons to achieve close to 1″ impact damage diameter. The greatest challenge in conducting these experiments was in the thick 4-mm and 6-mm coupons which did not display a predictable trend. Future work should focus on overcoming this challenge towards achieving a predictable impact damage size methodology.

Smart self-sensory carbon-based textile reinforced concrete structures for structural health monitoring

The goal of this study is to develop a structural health monitoring methodology for smart self-sensory carbon-based textile reinforced concrete elements. The self-sensory concept is based on measuring the electrical resistance change in the carbon roving reinforcement and by means of an engineering gage factor, correlating the relative electrical resistance change to an integral value of strain along the location of the roving. The concept of the nonlinear engineering gage factor that captures the unique micro-structural mechanism of the roving within the concrete matrix is demonstrated and validated. The estimated value of strain is compared to a theoretical value calculated by assuming a healthy state. The amount of discrepancy between the two strain values makes it possible to indicate and distinguish between the structural states. The study experimentally demonstrates the engineering gage factor concept and the structural health monitoring procedure by mechanically loading two textile reinforced concrete beams, one by a monotonic loading procedure and the other by a cyclic loading procedure. It is presented that the proposed structural health monitoring procedure succeeded in estimating the strain and in clearly distinguishing between the structural states.

Impedance-based structural health monitoring applied to steel fiber-reinforced concrete structures

The goal of the present contribution is to test the effectiveness of an impedance-based structural health monitoring (ISHM) methodology to detect damage in steel fiber-reinforced concrete structures. For this aim, a piezoceramic sensor coated with epoxy is used. A prismatic concrete specimen was evaluated for the case in which the damage condition was obtained from compression testing. The impedance signatures were measured both before and after the compression testing for the specimens placed inside an environmental chamber. An optimization procedure was applied to compensate for the temperature effects that appear on the impedance signatures. For illustration purposes, the coated piezoelectric transducer was initially bonded to the surface of an aluminum beam to detect damage induced by a loosening bolt. An impedance analyzer was used to measure impedance signatures in the frequency range of 40–70 kHz for the procedure with the aluminum beam and 50–80 kHz for the prismatic steel fiber concrete specimen. An in-house portable impedance meter (SySHM impedance meter) was used to measure the impedance signatures. The results encourage the use of the ISHM approach for health monitoring of steel fiber-reinforced concrete structures.

A railway tunnel structural monitoring methodology proposal for predictive maintenance

Structural health monitoring (SHM) on railway tunnels is increasingly demanding inspection methodologies based on full-field image-based techniques. The relevance of adequate inspection methodologies to signal potential problems that need increased attention is at the forefront of any monitoring system for problems that can disturb tunnel's integrity. Detecting defects is decisive for an efficient inspection, investigating tunnel status, and rescheduling maintenance operations, when and where required. This paper introduces a methodology design, based on acquisition of tunnel's 3D geometry and monitoring it through time to signal the possible emergence of structural defects and return the exact position where monitoring equipment will have to be deployed to. A demonstrator is built, and the applicability and accuracy of the system is assessed through implementation of a model prototype, carried out on a scaled tunnel model. At a secondary stage inspection, imposed geometrical changes and defects of the tunnel are monitored and characterised by an optical nondestructive inspection technique based on the digital image correlation (DIC). It enables the appraisal of structural integrity on the tunnel lining through displacement and strain fields acquisition. The chief advantage of the deployed inspection system, amongst others, lies on its robust functionality in both geometrical and structural integrity inspection aspects, contributing to the establishment of a robust SHM methodology for railway tunnels over time. In conclusion, experimental results are presented for evaluating scaled tunnel condition and monitoring the tunnel profile, and further DIC data are presented inferring displacement field progress of an introduced structural defect.

Strain-based health indicators for the structural health monitoring of stiffened composite panels

A common defect of composite-stiffened structures is the disbond at the interface between the two constituents (skin/stringer), as a result of inefficient manufacturing process or foreign object impacts in service. Generally, discontinuities within the volume of an elastic solid medium, subjected to mechanical load, cause anomalies on the strain field in the near vicinity of the discontinuity. Utilizing this observation, this work investigates the effect of artificially induced disbonds in the skin/stiffener interface of an aeronautical-grade generic element. A structural health monitoring methodology is developed, leveraging on numerically simulated strains along the stringer foot which aims to assess the health state of the panel as the size of the disbonds increases. The study is implemented using a parametric finite element model generating various disbond scenarios. Longitudinal strain values are acquired at the exact points where in reality actual fiber Bragg grating sensors will be located. Two types of commonly utilized strain-based health indicators are evaluated, and their drawbacks are revealed and discussed. A new health indicator is proposed that proves its capability to monitor growing disbonds while being both load- and baseline-independent.

Study of sensitivity for strain-based structural health monitoring

One of the available methodologies for structural health monitoring (SHM) is based on strain field pattern recognition where, through the use of sensors capable of measuring strain on discrete points and machine learning techniques, it is possible to detect a damage event. In this study, strain data from fiber optic sensors (FOS), in particular fiber Bragg gratings (FBG), acquired through two experiments are used: an aluminum beam with 32 FBGs and CFRP beam provided with 20 FBGs, which serves as the main wing’s structure of an unmanned aerial vehicle (UAV). Both structures were subjected to dynamic loading for a pristine condition and later, for artificially damaged conditions. In the experiments presented in this paper the beams were provided with different amounts of sensors which were removed one by one in order to analyze the sensitivity of the damage detection methodology based on PCA to a change in the number of sensors. The results demonstrated that there are few sensors that contribute mostly to the methodology’s performance, these sensors are validated to be the ones located near the analyzed damage condition. Therefore, this study is the first step into the development of methodologies of damage localization using strains.

Analysis of barely visible impact damage severity with ultrasonic guided Lamb waves

Barely visible impact damage is one of the most common types of damage in carbon-fibre-reinforced polymer composite structures. This article investigates the potential of using ultrasonic guided Lamb waves to characterise the through thickness severity of barely visible impact damage in thin carbon-fibre-reinforced polymer structures. In the first step, a laser Doppler vibrometer was used to capture the full damage interaction of the wavefield excited by a piezoelectric actuator. Damage-scattered wavefield for four different severities were studied to find the best parameters for characterising the severity of damage. To reduce the overall acquisition time and size of data collected using the laser Doppler vibrometer, the measured signals were reconstructed from a singular broadband chirp response using a post-processing algorithm. From the full wavefield analysis obtained at a wide range of toneburst frequencies, the results showed that barely visible impact damage severity could be characterised using ultrasonic guided Lamb waves and that the mode, dominant at lower frequencies, gave better results than the mode. In the second step, the parameters for characterising the damage severity were applied to a sparse network of transducers as an in-service structural health monitoring methodology. The damage was successfully detected and located. In addition, the transducer path close to the predicted damage location was utilised to successfully quantify the damage severity based on the proposed damage index.

On the Influence of Capillary-Based Structural Health Monitoring on Fatigue Crack Initiation and Propagation in Straight Lugs.

This paper addresses the influence on the fatigue life induced by the implementation of a capillary-based structural health monitoring methodology, patented under the name eSHM. It consists in integrating structurally small and pressurized capillaries into the component, so that when a fatigue crack breaches the capillary network, it results in a leak flow to the open atmosphere and loss of pressure in the galleries which is detected by a pressure sensor. The novelty of the proposed system resides in the opportunity to locate the capillary according to the designer’s need, as one resorts to additive manufacturing for the part production. However, the presence of these galleries in highly stressed regions raises concerns about crack initiation at the capillary itself and accelerated fatigue crack growth. This paper aims at the quantification of the influence the eSHM has on the fatigue behavior of the component and the determination whether this influence is significant or not. To that purpose, numerical simulations on a straight lug component, using the finite elements and eXtended Finite Elements Methods (XFEM), are performed. Various capillary sizes and shapes are assessed, so as to enable a general conclusion on the impact of the eSHM methodology in straight lugs.

Data interpretation framework integrating machine learning and pattern recognition for self-powered data-driven damage identification with harvested energy variations

Data mining methods have been widely used for structural health monitoring (SHM) and damage identification for analysis of continuous signals. Nonetheless, the applicability and effectiveness of these techniques cannot be guaranteed when dealing with discrete binary and incomplete/missing signals (i.e., not continuous in time). In this paper a novel data interpretation framework for SHM with noisy and incomplete signals, using a through-substrate self-powered sensing technology, is presented within the context of artificial intelligence (AI). AI methods, namely, machine learning and pattern recognition, were integrated within the data interpretation framework developed for use in a practical engineering problem: data-driven SHM of plate-like structures. Finite element simulations on an aircraft stabilizer wing and experimental vibration tests on a dynamically loaded plate were conducted to validate the proposed framework. Machine learning algorithms, including support vector machine, k-nearest neighbor, and artificial neural networks, were integrated within the developed learning framework for performance assessment of the monitored structures. Different levels of harvested energy were considered to evaluate the robustness of the SHM system with respect to such variations. Results demonstrate that the SHM methodology employing the proposed machine learning-based data interpretation framework is efficient and robust for damage detection with incomplete and sparse/missing binary signals, overcoming the notable issue of energy availability for smart damage identification platforms being used in structural/infrastructure and aerospace health monitoring. The present study aims to advance data mining and interpretation techniques in the SHM domain, promoting the practical application of machine learning and pattern recognition with incomplete and missing/sparse signals in smart cities and smart infrastructure monitoring.

An experimental study on the data-driven structural health monitoring of large wind turbine blades using a single accelerometer and actuator

Abstract The study aimed to investigate the performance of a Structural Health Monitoring (SHM) methodology based on use of a single accelerometer and single actuator in the detection and monitoring of the growth of the damage in the trailing edge of a wind turbine blade. The study used a data-driven vibration SHM, which is considered as a simple, nonparametric method for data compression and information extraction. The methodology found combinations of variables that describe major trends and fluctuations within the vibration response measured on the structure to create a reference state to which new observations were evaluated for damage diagnosis. The blade was artificially excited with an electromechanical actuator that introduced a mechanical impulse in the blade. The vibration responses were measured by accelerometers distributed along the trailing and leading edge of the wind turbine blade. The rationale behind this study was to investigate the combination of accelerometer/actuator location for damage detection sensitivity and damage progression when a single accelerometer and single actuator was used. The experimental study was conducted on an SSP 34 m wind turbine blade with and without introduced damage. Different damage sizes were also considered to evaluate the detectability of the damage. This study complements previous analyses in the same blade where studies on the effect of damage in modal parameters and a multiple sensor SHM technique were evaluated. The results demonstrated that the methodology was able to detect different damage sizes by using only one accelerometer. It was also demonstrated that damage detection and damage progression is affected by the accelerometer/actuator position but this effect is used to provide a rough information about damage location.

Structural damage detection based on improved frequency change ratio

Frequency Change Ratio (FCR) based damage detection methodology for structural health monitoring (SHM) is analyzed in detail. The effectiveness of damage localization using FCR for some slight damage cases and worse ones are studied on an asymmetric planar truss numerically. Disadvantages of damage detection using FCR in practical application are found and the reasons for the cases are discussed. To conquer the disadvantages of FCR, an Improved Frequency Change Ratio (IFCR) based damage detection method which takes the changes of mode shapes into account is proposed. Verification is done in some damage cases and the results reveal that IFCR can identify the damage more efficiently. Noisy cases are considered to assess the robustness of IFCR and results indicate that the proposed method can work well when the noise is not severe.

An estimate of the location of multiple delaminations on aeronautical CFRP plates using modal data inverse problem

With the increase in the use of composite materials, especially in the aeronautical industry, it is essential that a complete evaluation of the mechanical performance of such structures be undertaken, especially with regard to structural integrity. To assist in this task, structural health monitoring methodologies are employed in order to minimize time and maintenance costs, and errors arising principally from human factors, and which can occasionally result from the failure to properly inspect the aircrafts. This study addresses the use of an inverse method for delamination identification in carbon fiber reinforced polymers plates. First, the direct problem was modeled via a finite element method in order to obtain a faithful model that represented the real case studied. The inverse problem was solved by minimizing an objective function through genetic algorithms. Modal responses of delaminated plates are able to identify the possible location of multiple delaminations in laminated plates since the structural matrices are changed as a function of the induced damage. Numerical and experimental results showed excellent identification of small delaminations, reducing the initial search area by up to 96%, which can lead to savings in time and costs for the aeronautical industry.

Numerical evaluation of seismic-induced damages in masonry elements of historical arch viaduct

The main purpose of this research was to estimate the seismic response of the historical masonry arch viaduct located in Krakow (Southern Poland) subjected to the representative strong seismic event. For the numerical evaluation a 3D finite element model (FEM) of the masonry viaduct was assembling with the ABAQUS/Standard software program. To represent nonlinear behaviour of the masonry structural elements of the viaduct under the strong seismic event, a Barcelona Model (BM) was assumed as a constitutive material model. Three components of the registered strong seismic shock were applied as kinematic excitation acting in three directions. The results of the numerical simulation showed that evident nonlinear behaviour of the masonry structural elements of the viaduct was observed under the earthquake. The plastic strains as well as the tensile damage (cracking) were obtained in some zones of the walls of the structure after seismic shock. On the basis of the obtained results authors also discussed the methodology for structural health monitoring (SHM) as well as seismic retrofitting which could be dedicated for historical masonry arch viaducts and bridges.

Stress intensity factor monitoring under cyclic loading by digital image correlation

AbstractIn the present work, a methodology for structural health monitoring based on a combination of digital image correlation and an analytical elastic solution is presented. To this end, full‐field displacement around a crack tip in a CT sample made of 2024‐T351 Al alloy under cyclic loading was monitored at different load levels. An analytical solution based on Williams' model was used to evaluate the experimental value of the stress intensity factor (SIF) in a continuous fashion during cyclic loads. It was observed that by increasing the loading amplitude in the cyclic loading, the difference between nominal and experimental estimation of SIF increased due to the crack tip plasticity effect, which was not considered in the nominal evaluations. To consider the plasticity effect, Irwin's approach was employed. The results showed that the proposed method can successfully monitor the evolution of SIF of a sample under cyclic loading until the sudden fracture of the sample.