PSI - Issue 52

Yuhang Pan et al. / Procedia Structural Integrity 52 (2024) 699–708 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

700

2

challenging to detect with the current non-destructive inspection techniques during the maintenance), assessment, localization and prognosis throughout the life cycle of structures with minimal manual intervention (Qing et al. 2022). A great deal of research has been carried out on SHM methods for aerospace structures, and the most commonly used methods can be categorized into guided wave-based SHM (GWSHM) and vibration based SHM (VSHM) by the type of integrated sensor type (Kralovec and Schagerl 2020; Toh and Park 2020). GWSHM methods use the mounted piezoelectric (PZT) sensors to actuate and receive guided wave and detect damage by comparing signals obtained from the current state of structures with a baseline pristine state. The main advantage of GWSHM is its ability to detect the barely visible damage due to the high frequency excitation. Another advantage is that PZT sensors are light weight and easy to mount on or embed into structures. As such, many GWSHM methods have been developed for aerospace structures such as metal and composite in laboratory environment. Ricci et al. (Ricci et al. 2022) reviewed state-of-the-art of GWSHM methods for composited aerospace structures, which includes analytical models and numerical modes as well as their limitations. Yue et al. (Yue et al. 2021) presented a GWSHM approach for the damage detection and localization of flat plates and composite stiffened panel under different temperatures. Giannakeas et al. (Giannakeas et al. 2022) presented a similar framework and is validated by a large composite panel, and results showed that it can achieve both damage detection and localization under different temperatures. VSHM methods use accelerometers to detect damage by comparing changes in modal parameters before and after damage to structures, and these VSHM methods can be subdivided into natural frequency based (Negru et al. 2015), modes shape based (Huang et al. 2016.) and frequency response function (FRF) (Bandara et al. 2014; Nayyar et al. 2022) based by the type of modal parameters. As modal parameters are inherent to structures and insensitive to temperatures, VSHM methods are capable of detecting damage under different temperatures, which is one of the main shortcomings of the GWSHM methods that are temperature sensitive(Giannakeas et al. 2023; Ren et al. 2023). Zhang et al. (Zhang et al. 2018) proposed a natural frequency-based method to predict the location and the delamination sizes, and it was validated by the modal analysis of fibre reinforced polymer (FRP) composites with delamination. Khan et al. (Khan et al. 2019) reported a method that detects the delamination in composite laminates by using low-frequency vibration response. The low-frequency vibration response is effective for detecting the presence of damage which is also demonstrated by Minak et al. (Minak et al. 2010). Castellini et al.(Castellini and Marco Revel 2000) combined FRF data and neural network to detect and localize the delamination on composite panels. Although GWSHM and VSHM methods have been widely used in aeronautics, some aspects of these two methods have yet to be addressed. GWSHM methods are sensitive to environmental variables. For example, it is difficult to detect whether the change of signal is due to damage or temperature when temperature changes. In terms of VSHM. Its main challenge is the difficulty in detecting minute damage due to the insensitivity of modal parameters to small damage. In addition, modal parameters, especially mode shapes, are only sensitive to the local damage (Fan et al. 2011), and fail to provide a reliable results for SHM of large structures. Therefore, this study proposes a hybrid method combining GWSHM and VSHM, and aims to achieve a higher accuracy in damage detection and localization. In the following section, the experimental platform is firstly presented. Then, the proposed methodology that consists of feature extraction, model development and evaluation is explained, and corresponding results are presented and discussed. 2. Experimental set-up Based on the aim of this research, a hybrid system combining vibration and guided wave is developed, as shown in Fig.1. Four piezoelectric transducers (PZTs) utilized to actuate and receive guided wave, are evenly bonded to the aluminum plate, with a spacing of 120mm between each transducer. Two accelerometers and a force transducer are attached on the front and back side to get the modal properties of the plate, as shown in Fig.1 (b) and (c). The shaker is used to vibrate the plate, and the signal is captured by the Spider and Crystal Instruments EDM Software, the excitation signal is sweeping sine with the frequency from 5 to 2000Hz, as shown in Fig.1 (c). The guided wave acquisition system comprises the arbitrary generator, data acquisition, switching modules and the LASAR software, the actuation signal used in this research is chirp signal with frequency between 25KHz and 600KHz, as shown in Fig.1 (d).