An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

Structural health monitoring for lightweight complex composite structures is being investigated in this paper with a data-driven deep learning approach to facilitate automated learning of the map of transformed signal features to damage classes. Towards this, a series of acoustic emission (AE) based laboratory experiments have been carried out on a composite sample using a piezoelectric AE sensor network. The registered time-domain AE signals from the assigned sensor networks on the composite panel are processed with the continuous wavelet transform to extract time-frequency scalograms. A convolutional neural network based deep learning architecture is proposed to automatically extract the discrete damage features from the scalogram images and use them to classify damage-source regions in the composite panel. The proposed deep-learning approach has shown an effective damage monitoring potential with high training, validation and test accuracy for unseen datasets as well as for entirely new neighboring damage datasets. Further, the proposed network is trained, validated and tested only for the peak-signal data extracted from the raw AE data. The application of peak-signal scalogram data has shown a significant improvement in damage-source classification performance with high training, validation and test accuracy.

Acoustic emission data based deep learning approach for classification and detection of damage-sources in a composite panel

In this paper, a passive structural health monitoring (SHM) method capable of detecting the presence of damage in carbon fibre/epoxy composite plates is developed. The method requires the measurement of strains from the considered structure, which are used to set up, train, and test artificial neural networks (ANNs). At the end of the training phase, the networks find correlations between the given strains, which represent the ‘fingerprint’ of the structure under investigation. Changes in the distribution of these strains is captured by assessing differences in the previously identified strain correlations. If any cause generates damage that alters the strain distribution, this is considered as a reason for further detailed structural inspection. The novelty of the strain algorithm comes from its independence from both the choice of material and the loading condition. It does not require the prior knowledge of material properties based on stress-strain relationships and, as the strain correlations represent the structure and its mechanical behaviour, they are valid for the full range of operating loads. An implementation of such approach is herein presented based on the usage of a distributed optical fibre sensor that allows to obtain strain measurement with an incredibly high resolution.

/pdf/damage-detection-in-composites-by-artificial-neural-networks-2bun2o15ox.pdf

Damage Detection in Composites By Artificial Neural Networks Trained By Using in Situ Distributed Strains

Fibre-reinforced polymer (FRP) composite materials are widely used in different branches of industry, especially in aerospace, owing to their low mass, high strength and stiffness, and good fatigue and corrosion resistance. However, these materials are prone to the impact damage. Especially dangerous are barely visible impact faults, since it is difficult to detect them. If left unrepaired, they can lead to collapse of the whole structure. Hence, a continuous monitoring for loads and possible impact faults in these structures is crucial. Traditionally, this is realized via surface-mounted sensor technologies. However, smart structures with internally embedded sensors offer several advantages — sensor protection from the environment, better coupling to the structure, and no disruption of surface geometry, which is essential for aerodynamic elements, also allowing monitoring in the real time without stopping their operations. The most popular existing smart structural solutions — piezoelectric sensor networks and fibre optics, are reviewed along with other, less common sensor choices. This review also covers the limitations associated with sensor embedment, whose addressing would bring the society to a more reliable, cheaper, and efficient maintenance of transportation and infrastructure.

Smart Composite Structures with Embedded Sensors for Load and Damage Monitoring – A Review

ABSTRACT Composite materials have been widely used in many industries due to their excellent mechanical properties. It is difficult to analyze the integrity and durability of composite structures because of their own characteristics and the complexity of load and environments. Structural health monitoring (SHM) based on built-in sensor networks has been widely evaluated as a method to improve the safety and reliability of composite structures and reduce the operational cost. With the rapid development of machine learning, a large number of machine learning algorithms have been applied in many disciplines, and also are being applied in the field of SHM to avoid the limitations resulting from the need of physical models. In this paper, the damage monitoring technologies often used for composite structures are briefly outlined, and the applications of machine learning in damage monitoring of composite structures are concisely reviewed. Then, challenges and solutions for quantitative damage monitoring of composite structures based on machine learning are discussed, focusing on the complete acquisition of monitoring data, deep analysis of the correlation between sensor signal eigenvalues and composite structure states, and quantitative intelligent identification of composite delamination damage. Finally, the development trend of machine learning-based SHM for composite structures is discussed.

https://www.tandfonline.com/doi/pdf/10.1080/19475411.2022.2054878?needAccess=true

Machine Learning Based Quantitative Damage Monitoring of Composite Structure

In-flight and wireless damage detection in a UAV composite wing using fiber optic sensors and strain field pattern recognition

Composite materials have been extensively used on new aircraft airframes because of their advantages over metallic materials. This represents a difficulty for damage detection, a vital task for safety on the aerospace industry, as most nondestructive testing techniques are not effective on these materials since those usually present internal failures like delaminations which are difficult to detect. A miniaturized strain acquisition and wireless transmission system is presented alongside with a novel technique for structural behavior assessment, based on the use of Fiber Bragg Gratings to measure strains and non-supervised classification techniques to recognize different operational conditions. Operational tests were performed on an Unmanned Aerial Vehicles wing's beam, made of composite materials with the sensors embedded during its manufacturing. Strain measurements were processed using an Optimal Baseline Selection methodology. The tests performed proved the system's capability to identify and separate different operational conditions for a healthy structure, based on the analysis of its strain fields. The implementation of this methodologies can lead to perform real-time damage detection on aerospace complex structures made of composite materials.

Alejandro Carvajal-Castrillón

Papers

In-flight and wireless damage detection in a UAV composite wing using fiber optic sensors and strain field pattern recognition

Structural health monitoring on an unmanned aerial vehicle wing's beam based on fiber Bragg gratings and pattern recognition techniques