Machine learning techniques such as deep learning have already been successfully applied in Structural Health Monitoring (SHM) for damage localization using Ultrasonic Guided Waves (UGW) at various temperatures. However, a common issue arises due to the time-consuming nature of collecting guided wave measurements at different temperatures, resulting in an insufficient amount of training data. Since SHM systems are predominantly employed in sensitive structures, there is a significant interest in utilizing methods and algorithms that are transparent and comprehensible. In this study, a method is presented to augment feature data by generating a large number of training features from a relatively limited set of measurements. In addition, robustness to environmental changes, e.g., temperature fluctuations, is improved. This is achieved by utilizing a known temperature compensation method called temperature scaling to determine the function of signal features as a function of temperature. These functions can then be used for data generation. To gain a better understanding of how the damage localization predictions are made, a known explainable neural network (XANN) architecture is employed and trained with the generated data. The trained XANN model was then used to examine and validate the artificially generated signal features and to improve the augmentation process. The presented method demonstrates a significant increase in the number of training data points. Furthermore, the use of the XANN architecture as a predictor model enables a deeper interpretation of the prediction methods employed by the network.

For the safe operation of vehicles and full utilization of lightweight materials, assurance of structural integrity is a prerequisite at all times. Structural health monitoring with permanently installed transducers offers a great advantage for primary load-bearing structures of all means of transportation and other safety-relevant components such as hydrogen tanks: allowing damage detection during operation. One means to detect internal defects is the method of guided ultrasonic waves (GUWs), which can be generated and recorded by piezoelectric transducers. GUWs propagate along the elongated dimension of a structure, and a transducer network can completely cover and monitor structures. Defects can alter the signal along affected paths and allow for their detection. However, a challenge and obstacle for the application of such a testing technique in the service of means of transportation is the large influence of temperatures. These influences are difficult to distinguish from the effect of defects. One approach to overcome this difficulty is the “continuous baseline update”. Recurrence quantification analysis is tested and compared to established features as a new approach to “continuous baseline update” in this paper. Publicly available GUW data (http://www.openguidedwaves.de/) recorded under varying temperature conditions have been used to show how the methods perform. They reliably separate temperature and damage effects, while the recurrence quantification analysis yields the best results.

Automated damage detection in carbon fibre and fibre metal laminates is still a challenge. Impact damage is typically not visible from the outside. Different measuring and analysis methods are available to detect hidden damage, e.g., delamination or cracks. Examples are X-ray computer tomography and methods based on guided ultrasonic waves (GUW). All measuring techniques are characterised by high-dimensional sensor data—in the case of GUW, a set of time-resolved signals as a response to an actuated stimulus. We present a simple but powerful two-level method that reduces the input data (time-resolved sensor signals) significantly through a signal feature selection computation, which can then be applied to a damage predictor function. Besides multi-path sensing and analysis, the novelty of this work is a feed-forward ANN with low complexity which is used to implement the predictor function: it combines a classifier and a spatial regression model.

In this work, we present the full-field measurement of guided ultrasonic waves (GUWs) on technical surfaces using lensless digital holography. The setup works with a pulsed coherent laser illumination and a complementary metal-oxide semiconductor (CMOS) sensor without using any imaging optics. It provides highly resolved single-shot height measurements featuring the deformation introduced to the material by ultrasound excitation via piezoelectric actuators. For the evaluation of the resulting height distribution, we propose the multidirectional structure function offering pixel-precise determination of the wavelength. This ability is used to carry out a comparison between two aluminum plates of different thicknesses. Here, the resulting difference in wavelengths reflects the change in the thickness and is in agreement with simulated results for the two specimens, qualifying the proposed method as a detector for GUWs.

Guided ultrasonic wave (GUW) analysis is a well-investigated method for structural health monitoring (SHM) applications. For plate-like structures, the pitch-catch technique is a popular choice since it offers the possibility to investigate a large area with a small number of sensors. This method requires a large amount of data to be analyzed to detect and localize damage, with the consequence that, besides the presence of damage, environmental influences like temperature and load will also change the GUW signals. In addition, the location, size, and type of the damage will result in different changes in the GUW signals. Data-driven methods require sufficient data and therefore require data augmentation. In order to get closer to this goal, this study aims to demonstrate the conversion of GUW signals measured with an air-coupled measurement system (ACMS) into signals measured with piezoelectric wafer active sensors (PWAS). This would allow the fast measurement of GUW data with ACMS at different positions of a plate-like specimen and translate it to a surface-bonded PWAS signal without the time-consuming process of transducer mounting. In this study, it is assumed that the measurement methods are not independent of each other when they are measured at the same position. To obtain the transform function from ACMS to PWAS, GUW signals were measured both with ACMS and PWAS for different positions of artificial damage. Since both signal classes are physically dependent, it should be possible to determine the transform function with machine learning (ML) methods. As input, the ACMS time-dependent signal or signal features are used, while the PWAS signals serve as labels for the training process. We are evaluating different ML-based transform model architectures with respect to their suitability for signal or signal feature transformation, e.g., ANN, CNN, and LSTM-based networks, with a particular focus on autoencoders.