Nonlinear damage detection in composite structures using bispectral analysis

In composite structures, damages are often invisible from the surface and can grow to reach a critical size, potentially causing catastrophic failure of the entire structure. Thus safe operation of these structures requires careful monitoring of the initiation and growth of such defects. Ultrasonic methods using guided waves offer a reliable and cost-effective method for structural health monitoring in advanced structures. Guided waves allow for long monitoring ranges and are very sensitive to defects within their propagation path. In this work, the relevant properties of guided Lamb waves for damage detection in composite structures are investigated. An efficient numerical approach is used to determine their dispersion characteristics, and these results are compared to those from laboratory experiments. The experiments are based on a pitch-catch method, in which a pair of movable transducers is placed on one surface of the structure to induce and detect guided Lamb waves. The specific cases considered include an aluminum plate and an aluminum honeycomb sandwich panel with woven composite face sheets. In addition, a disbond of the interface between one of the face sheets and the honeycomb core of the sandwich panel is also considered, and the dispersion characteristics of the two resultant waveguides are determined. Good agreement between numerical and experimental dispersion results is found, and suggestions on the applicability of the pitch-catch system for structural health monitoring are made.

Deep transfer learning-based damage detection of composite structures by fusing monitoring data with physical mechanism

Woven fabric composite structures are applied in a wide range of industrial applications. Composite structures are vulnerable to damage from working in complex conditions and environments, which threatens the safety of the in-service structure. Damage detection based on Lamb waves is one of the most promising structural health monitoring (SHM) techniques for composite materials. In this paper, based on guided Lamb waves, a lightweight deep-learning approach is proposed for identifying damaged regions in woven fabric composite structures. The designed deep neural networks are built using group convolution and depthwise separated convolution, which can reduce the parameters considerably. The input of this model is a multi-channel matrix transformed by a one-dimensional guided wave signal. In addition, channel shuffling is introduced to increase the interaction between features, and a multi-head self-attention module is designed to increase the model’s global modeling capabilities. The relevant experimental results show that the proposed SHM approach can achieve a recognition accuracy of 100% after only eight epochs of training, and the proposed LCANet has only 4.10% of the parameters of contrastive SHM methods, which further validates the effectiveness and reliability of the proposed method.

In this paper, an improvement of the elliptical triangulation method for damage detection using Lamb waves is presented. The damage is the main cause of structural failure and often occurs on structures. In order to avoid sudden failure, a special attention was given in the past decades to the damage detection in structures. In order to obtain efficient damage detection techniques, the structural health monitoring has been the main research topic of many scientists worldwide. The elliptical triangulation method, proposed in this paper, is a non-destructive method based on measurement of Lamb waves. This method, through the calculation of the time of flight of the signals and the actuator-sensor positioning, allows to identify position and dimension of the damage. The application of the method to the metallic structures and to the composite material structures is presented in this paper. The complexity connected with the uncertainty of the waves’ propagation speed due to the anisotropy of the composite materials has been explored through an iterative approach. The initialization of the wave propagation speed at first tentative iteration is the key issue for the convergence of the method. Seven different conditions were used to validate the method on both metallic and composite structures combining two damage shapes, two damage dimensions (effective damaged area), and three different positions. Upon evaluating the effectiveness, the method has been applied at two composite panels in order to detect by test the post-impact damages. Tests results have been compared with the numerical ones. The feasibility of the elliptical triangulation method to detect the damage (evaluating the damage position and area) has been proved using the ultrasonic C-Scan.

Many composite materials and structures are susceptible to defects, which can significantly reduce the strength of structures and may grow to failure. To avoid the catastrophic failure of structures, development of a reliable method of structural health monitoring is one of the most important keys in maintaining the integrity and safety of structures. Dynamic response-based damage detection offers a simple procedure as an alternative to the conventional nondestructive evaluation techniques. However, this technique depends on the quality of measured data for its identification accuracy. In this article, experimental aspects of dynamic response-based damage detection technique on carbon/ epoxy composites are addressed. Smart piezoelectric materials are used as sensors or actuators to acquire the curvature modes of structures. These materials are surface-bonded to the beams. An impulse hammer is used as an actuating source as well. Four types of damage detection algorithms are evaluated for several possible damage configurations with two different excitation sources. The quality of damage identification with the four different detection algorithms is discussed. These experimental damage identification techniques using curvature modes and piezoelectric materials can be effectively used in damage detection and health monitoring of composite structures.

In this paper, the temperature scaling approach (TS) for temperature compensation of guided ultrasonic waves (GUW) is presented. In many previous works, it could be observed that the amplitudes of the GUW change with temperature. In the TS approach, it is assumed that for each transducer configuration several functions \(M\left( f,T\right) \) exist that describe the change of the maximum amplitude depending on the frequency f and the temperature T. It is also assumed that several functions \(F\left( f,T\right) \) exist that describe the changes of the full width at half maximum (FWHM) from the amplitude containing the maximum value of the measurement. Furthermore, it is assumed that the functions \(M\left( f,T\right) \) and \(F\left( f,T\right) \) also depend on the medium in which the GUW propagate, so that a change of the medium results in a change of the functions. This means that these functions can be used for damage detection. GUW data from the http://openguidedwaves.de/ database were analyzed with the TS approach. Data from a carbon fiber reinforced polymer plate with a temperature range from 20 \(^\circ \)C to 60 \(^\circ \)C were analyzed for different actuator-sensor paths and compared with position-dependent artificial defects. To determine the functions \(M\left( f,T\right) \) and \(F\left( f,T\right) \), polynomials were fitted to the measured values of the maxima and full FWHM. The TS analysis showed a detect ability of the artificial defects over the entire temperature range.KeywordsSHMTemperature compensatingSignal processingData analysisEmbedded systems

Data-driven methods in structural health monitoring (SHM) is gaining popularity due to recent technological advancements in sensors, as well as high-speed internet and cloud-based computation. Since the introduction of deep learning (DL) in civil engineering, particularly in SHM, this emerging and promising tool has attracted significant attention among researchers. The main goal of this paper is to review the latest publications in SHM using emerging DL-based methods and provide readers with an overall understanding of various SHM applications. After a brief introduction, an overview of various DL methods (e.g., deep neural networks, transfer learning, etc.) is presented. The procedure and application of vibration-based, vision-based monitoring, along with some of the recent technologies used for SHM, such as sensors, unmanned aerial vehicles (UAVs), etc. are discussed. The review concludes with prospects and potential limitations of DL-based methods in SHM applications.

In civil engineering, health monitoring of structure have been treated as an important subject. Health monitoring by using experimental measurements such as strain, acceleration, velocity, displacement, rotation and other parameters has been used for detecting changes, which may indicate damage or degradation. In recent, the applications of health monitoring of large structure have been facilitated by the developments in measurement, communications and computational technology gradually. Especially, large civil structures as bridges are damaged by fatigue loads or unexpected load such as earthquakes and finally reach to failure. Thus, a structural health monitoring and damage detection methods have been developed and attracted a great deal of public attention in recent decades. Actually many research efforts have been conducted to develop a health monitoring and damage detection techniques of existing civil structures. However, these methods are needed to improve to apply to a real civil structure. The purpose of this study is to develop a practical instrument to identify dynamic parameters of real civil structural systems.

Identification of impact and its resulting damage is a complex inverse inferential procedure. Estimating the magnitude of an impact from sensor data is challenging and still needs to be fully understood. Furthermore, only a few studies have solved practical impact identification problems relevant to real-world structures with different geometries and structural inhomogeneities (e.g. variable thicknesses and stiffeners). Accordingly, the current research presents an impact categorisation analysis using passive measurements from piezoelectric (PZT) sensors for a geometrically complex composite aircraft part. Intermediate-mass impact tests were performed on a stiffened thermoplastic composite plate to generate signals representing various impact scenarios that may occur during the lifetime of an aircraft. The impacts were applied at critical locations, including the mid-bay, on top of the stringer, and at different thicknesses. This study investigated the effect of structural features on wave propagation in composite structures. The experimental results demonstrate that these features significantly change the wave field and, consequently, affect sensor readings. To differentiate between pristine and damaged states, an impact signal categorisation approach was employed. This analysis shows that a passive sensing system can categorise the impact using the transmitted energy from a time-domain signal. However, the spectral energy values showed low efficiency in categorising impacts. Overall, the results highlight the importance of considering structural features when deploying sensor systems for structural health monitoring and provide valuable insights into the effectiveness of impact identification and damage detection in realistic composite structures.

Implementation of heterodyning effect for monitoring the health of adhesively bonded and fastened composite joints

In pipeline maintenance, the advent of cloud computing and Internet of Things technologies enable effective pipeline management. Practically, it is impossible to provide fast-response, full-scale, multiterrain, and spatial-level monitoring of pipelines by sparse sensors. In this article, a fitting method for online pipeline structural health monitoring (SHM) is presented. The contribution includes three aspects. First, a new pipeline SHM method based on spatial deformation fitting is proposed, which is useful for three-dimensional full-scale assessment and multiterrain applications with the sparse sensor array. There are no strict restrictions on the number and arrangement of sensors, and no finite-element-model prior information is needed. The monitoring results can be delivered to multiterminals. Second, a rational cubic spline-based fitting algorithm mainly implemented by cloud computing is provided for rapid response. It features three-time interpolation fitting and one-time coordinate transformation. The value correspondence between the measured attitude angles and the fitted deformation is dexterously established. The fitted curve is locally adjustable. Third, the fitting error on deformation is analyzed, and the effectiveness of the method is tested and verified. Reasonable parameter values for the sensor quantity and the fitting spline are suggested.

The emerging nanotechnology has ushered a new avenue to tailor conventional composite materials. In this study, a new kind of nano-engineered composite material has been developed with a self-sensing capability for guided ultrasonic waves (GUWs)- based structural health monitoring (SHM). In the composites, a graphene-based selfsensing network is designed, developed and “dispersed” in the epoxy matrix of conventional glass fibre-reinforced composites. The nano-structure of the graphene conductive network is optimized, via which the quantum tunnelling effect formed in the network is linked to material elastic deformation when GUWs traverse the composites, via measuring the changes in the conductivity of the dispersed network. Measurement can be implemented at any position of the composites, owing to the “dispersed” sensing capability of the composites, and this renders a possibility to monitor GUW propagation throughout the entire composite structures at any desired position of the structures, while without a need to use any additional sensor externally attached to or internally embedded in the composites. Experimental validation is conducted, in which GUWs that are self-sensed by the composites are compared with the counterpart signals obtained by conventional sensors (piezoelectric wafers), to observe that the developed composites can be responsive to GUWs with frequencies in an ultrasonic regime up to several hundreds of kilohertz, with no prominent discrepancy compared with conventional piezoelectric wafers. This study has spotlighted a new breed of composites with a capacity of self-sensing and self-diagnosis towards online health monitoring of composite materials and structures

A methodology for sensor number and placement optimization for vibration-based damage detection of composite structures under model uncertainty

This paper presents an overview of current research and development in the field of sensor placement methods in structural health monitoring. Due to the limitations of equipment facilities and cost, the number of sensors to be installed in a structure is relatively few. In many cases, three are always degrees of freedoms in a structure are not easily accessible, or eventually inaccessible. Moreover, actual structures are all continuous and have essentially infinite degrees of freedoms, therefore, it is impossible to instrument on all the degrees of freedoms. The methods of sensor placement are mainly dealing with accelerometers for global dynamic testing, or structural health monitoring. Three major concerns are discussed, the necessary number of sensors to be installed, where to deploy these sensors, and how to evaluate the effectiveness of these deployed sensors. Finally, a new loading dependent sensor placement method is proposed, and verified by the experimental data of a simple cantilever beam structure.

Unsupervised learning methods are effective and suitable tools for damage detection. The main reason for the popularity of these methods in structural health monitoring originates from the fact that the process of learning can be implemented by information of the only normal condition called training data. In contrast, supervised learning methods require information of both normal and current conditions for the process of interest. Because civil engineering structures are expensive and complex, it is not reasonable and economical to impose intentional damage on providing training data. Hence, it is not simple to directly exploit supervised learning techniques in structural health monitoring. To deal with this limitation, this article proposes a novel two-level strategy including three algorithms for using the concepts of both unsupervised learning and supervised learning. The major contribution of this strategy is to consider supervised learning as a validation tool for damage detection. First, the results of damage detection are obtained from two unsupervised learning methods developed by Mahalanobis squared distance and a deep autoencoder neural network in the first two algorithms of the proposed strategy. The main objective is to separate accurate and confusing results of damage detection based on Type I and Type II errors. Second, the confusing results are fed into the third algorithm to train a classifier and compute their classification margins for making the final decision and validating damage detection. The effectiveness and applicability of the proposed strategy are assessed by a numerical concrete beam and an experimental laboratory frame. Results show that this strategy with the aid of the Naive Bayes classifier enables the unsupervised learning methods to make accurate decisions.