Incipient Structural Damage Research Articles

Machine-Learning-Based Methods for Acoustic Emission Testing: A Review

Acoustic emission is a nondestructive control technique as it does not involve any input of energy into the materials. It is based on the acquisition of ultrasonic signals spontaneously emitted by a material under stress due to irreversible phenomena such as damage, microcracking, degradation, and corrosion. It is a dynamic and passive-receptive technique that analyzes the ultrasonic pulses emitted by a crack when it is generated. This technique allows for an early diagnosis of incipient structural damage by capturing the precursor signals of the fracture. Recently, the scientific community is making extensive use of methodologies based on machine learning: the use of machine learning makes a machine capable of receiving a series of data, modifying the algorithms as they receive information on what they are processing. In this way, the machine can learn without being explicitly programmed, and this implies a huge use of data and an efficient algorithm to adapt. This review described the methodologies for the implementation of the acoustic emission (AE) technique in the evaluation of the conditions and in the monitoring of materials and structures. The latest research products were also analyzed in the development of new methodologies based on machine learning for the detection and localization of damage for the characterization of the fracture and the prediction of the failure mode. The work carried out highlighted the strong use of these methods, which confirms the extreme usefulness of these techniques in identifying structural damage in scenarios heavily contaminated by residual noise.

Two-dimensional scattering features of the mixed second harmonic A0 mode Lamb waves for incipient damage localization

Inspection of incipient structural damage is of great significance. Existing techniques based on nonlinear guided waves have shown great promise for the detection of incipient damage due to material microstructure changes, but with only limited success for damage localization, which is technically more challenging. Our previous work uncovered the existence of the second harmonic A0 mode Lamb waves in a PZT-driven system, as a result of the mixing of the primary A0 and S0 mode Lamb waves, thus pointing at the possibility of performing damage localization through tuning the size of the wave mixing zone. In the present study, a two-dimensional incipient damage localization method is proposed based on this newly discovered wave phenomenon. By visualizing the nonlinear wave field, damage-induced nonlinear wave scattering features are first investigated. A dedicated localization algorithm is then proposed and evaluated. Numerical results show that the energy of the scattered nonlinear wave is mainly confined to a narrow region along the actuator-damage path, the spatial variation of which can be approximated by a simple Gaussian function. Embedding this information into the proposed localization strategy, damage localization can be achieved using a simple physical system. Experiments are finally carried out to validate the 2nd A0 wave scattering features and the proposed damage localization method.

Convolutional neural network and impedance-based SHM applied to damage detection

The impedance-based structural health monitoring technique uses measured signatures changes to identify incipient damages in structures. The purpose is to perform a correlation of these changes with the physical phenomena. However, since electromechanical coupling exists, some environmental influences such as temperature changes may lead to false decision regarding the condition of the structure. As a result, innovative machine learning tools have been extensively investigated to avoid errors in structural prognosis and, in this sense, recent applications of convolutional neural networks (CNN) have emerged within the scope of SHM research, focusing mainly on vibration analysis. However, studies that aim to combine neural architectures with intelligent materials for structural monitoring purposes have been poorly evaluated. Consequently, its integration with the electromechanical impedance method is still considered as being a new application of CNN. Thus, in order to contribute to the SHM area, this work presents a combination of the CNN architecture and the EMI methodology. In the present contribution, three aluminum beams subjected to three different steady temperature levels (0 °C, 10 °C and 20 °C) were studied. For this aim, a test chamber was used for humidity and temperature control. Artificial damages such as mass addition were taken into account so that impedance signatures related to both pristine and damaged conditions can be analyzed. Thus, a one-dimensional Convolutional Neural Network (1D CNN) was designed, trained and used for damage prediction purposes. In this context, a temperature robust model that is able to identify damage independently of environmental condition was developed.

A hybrid structural health monitoring technique for detection of subtle structural damage

There is greater significance in identifying the incipient damages in structures at the time of their initiation as timely rectification of these minor incipient cracks can save huge maintenance cost. However, the change in the global dynamic characteristics of a structure due to these subtle damages are insignificant enough to detect using the majority of the current damage diagnostic techniques. Keeping this in view, we propose a hybrid damage diagnostic technique for detection of minor incipient damages in the structures. In the proposed automated hybrid algorithm, the raw dynamic signatures obtained from the structure are decomposed to uni-modal signals and the dynamic signature are reconstructed by identifying and combining only the uni-modal signals altered by the minor incipient damage. We use these reconstructed signals for damage diagnostics using ARMAX model. Numerical simulation studies are carried out to investigate and evaluate the proposed hybrid damage diagnostic algorithm and their capability in identifying minor/incipient damage with noisy measurements. Finally, experimental studies on a beam are also presented to compliment the numerical simulations in order to demonstrate the practical application of the proposed algorithm.

Design of nonlinear-Lamb-wave-based structural health monitoring systems with mitigated adhesive nonlinearity

Structural health monitoring (SHM) techniques with nonlinear Lamb waves offer the possibility of detecting incipient damages through the second harmonic generation. Particularly, the primary S0-secondary S0 Lamb wave mode pair in a weakly nonlinear plate is promising for SHM due to its appealing characteristics such as the cumulative second harmonic generation and flexible frequency selection. In a practical PZT-actuated SHM system, apart from the damage-related material nonlinearity of the plate (MNP), the adhesive nonlinearity (AN) is also proven to be a non-negligible nonlinear source, which may jeopardize damage diagnosis if not properly apprehended. By combining a nonlinear shear-lag model and the normal mode expansion method, a theoretical model is proposed in this work, which allows the investigation and the comparison of these two types of important nonlinearities in the system (AN and MNP). With the developed model, optimized nonlinear Lamb wave-based SHM systems can be designed with the mitigated influence of the undesired AN on the system. Finite element (FE) simulations are carried out to validate the proposed model and the effectiveness of the optimized systems through a tactical adjustment of the different nonlinear sources. The designed optimized system is verified by experiments, in comparison with an arbitrarily chosen one. The dominance level of the nonlinear sources in the two systems is evaluated through changing the MNP, which is realized by a thermal aging treatment. Both FE and experimental results demonstrate that the proposed model can effectively guide the design of SHM systems to mitigate AN so that the MNP can be well measured and further used as an indicator of the incipient structural damage.

Method to Enhance the Frequency Readability for Detecting Incipient Structural Damage

Vibration-based damage detection methods able to recognize cracks in the early state require accurate frequency evaluation in order to identify small modal changes. Because of higher order, the natural frequencies are rapidly damped, time history provides insufficient information if standard frequency estimations are performed. This paper introduces a method of overlaying multiple spectra achieved by iterative truncation of the original signal. In this way, the spectral bin number becomes much higher and the identification of real natural frequencies is easier. Also, an original damage detection algorithm based on the frequency shift is described that considers the first six bending vibration modes occurred due to damage. These are in direct relation to the modal strain energy stored at the damage location. The effectiveness of the detection method is proved using real structures in some damage scenarios. Thus, it was shown that even if a crack reduces the beam’s transversal cross section by only 6%, it can be clearly detected.

Experimental Study on Damage Detection in Timber Specimens Based on an Electromechanical Impedance Technique and RMSD-Based Mahalanobis Distance.

In the electromechanical impedance (EMI) method, the PZT patch performs the functions of both sensor and exciter. Due to the high frequency actuation and non-model based characteristics, the EMI method can be utilized to detect incipient structural damage. In recent years EMI techniques have been widely applied to monitor the health status of concrete and steel materials, however, studies on application to timber are limited. This paper will explore the feasibility of using the EMI technique for damage detection in timber specimens. In addition, the conventional damage index, namely root mean square deviation (RMSD) is employed to evaluate the level of damage. On that basis, a new damage index, Mahalanobis distance based on RMSD, is proposed to evaluate the damage severity of timber specimens. Experimental studies are implemented to detect notch and hole damage in the timber specimens. Experimental results verify the availability and robustness of the proposed damage index and its superiority over the RMSD indexes.

Achieving selective interrogation and sub-wavelength resolution in thin plates with embedded metamaterial acoustic lenses

In this study, we present an approach to ultrasonic beam-forming and high resolution identification of acoustic sources having critical implications for applications such as structural health monitoring. The proposed concept is based on the design of dynamically tailored structural elements via embedded acoustic metamaterial lenses. This approach provides a completely new alternative to conventional phased-array technology enabling the formation of steerable and collimated (or focused) ultrasonic beams by exploiting a single transducer. Numerical results show that the ultrasonic beam can be steered by simply tuning the frequency of the excitation. Also, the embedded lens can be designed to achieve sub-wavelength resolution to clustered acoustic sources, which is a typical scenario encountered in incipient structural damage.

A coupled approach for damage detection of framed structures using piezoelectric signature

High-frequency electro-mechanical impedance (EMI) signature is very sensitive to the local incipient damages in structures. However, extensive information on the nature of damages may not be available via EMI technique in its conventional non-model-based form. On the other hand, there is also little analytical work on the vibration modes of complex structures at ultrasonic frequencies. In this paper, a coupled approach combining EMI technique and a reverberation matrix method is proposed to quantitatively correlate damages in framed structures with high-frequency signature for structural health monitoring. The structural members are modeled as Timoshenko beams for flexural motion and as the classical longitudinal rods for axial motion. The PZT wafers, which are bonded to the beams, are also treated as one-dimensional in an axial vibration. A shear lag model is adopted to simulate the interfacial bonding between PZT patches and the host beam. An analytical expression for impedance (or admittance) related to response of the coupled model of PZT patch–bonding layer–host frame system is derived for the first time. Comparison study is presented with other established methods and theories. Based on this analysis, EMI signatures are extracted to identify the damages in framed structures theoretically.

Wave propagation approach for NDE using surface bonded piezoceramics

This paper presents a wave propagation-based technique to detect and localize incipient damages in structures and machines. The technique utilizes arrays of surface-bonded piezoceramic (PZT) patches, which serve as transmitters and receivers of elastic waves through the monitored component. The operating frequency is maintained in the order of 100–150 kHz to enable high sensitivity in damage detection. The resulting frequency transfer functions facilitate the identification of the damaged region in the structure. The presented technique, which utilizes simple, economical and commercially available hardware and sensors, can be easily employed for real-time and online monitoring of critical structures, such as machine parts and aircraft components.

Structural Health Monitoring With Piezoelectric Active Sensors

A technology for non-intrusive real-time structural health monitoring using piezoelectric active sensors is presented. The approach is based on monitoring variations of the coupled electromechanical impedance of piezoelectric patches bonded to metallic structures in high-frequency bands. In each of these applications, a single piezoelectric element is used as both an actuator and a sensor. The resulting electromechanical coupling makes the frequency-dependent electric impedance spectrum of the PZT sensor a good mapping of the underlying structure’s acoustic signature. Moreover, incipient structural damage can be indicated by deviations of this signature from its original baseline pattern. Unique features of this technology include its high sensitivity to structural damage, non-intrusiveness to the host structure, and low cost of implementation. These features have potential for enabling on-board damage monitoring of critical or inaccessible aerospace structures and components, such as aircraft wing joints, and both internal and external jet engine components. Several exploratory applications will be discussed.

A Rational Definition of Yield Strength

Explicitly or implicitly the yield strength of a material is often used as a measure of incipient structural damage. With the yield strength determined by conventional methods, however, it cannot be said in general for two structural elements geometrically alike but of different materials that similar loads, producing maximum stresses equal to the yield strengths in the two cases, are simply related to the yield strengths. A definition of yield strength is proposed in this paper which often has the advantage that, for geometrically similar structures of different materials, loads producing maximum stresses equal to the yield strength are proportional to the yield strength.

A Rational Definition of Yield Strength

Abstract Explicitly or implicitly the yield strength of a material is often used as a measure of incipient structural damage. With the yield strength determined by conventional methods, however, it cannot be said in general for two structural elements geometrically alike but of different materials that similar loads, producing maximum stresses equal to the yield strengths in the two cases, are simply related to the yield strengths. A definition of yield strength is proposed in this paper which often has the advantage that, for geometrically similar structures of different materials, loads producing maximum stresses equal to the yield strength are proportional to the yield strength.