Search

Impact damage localization is crucial in aircraft structural health monitoring. The integration of ultrasonic-guided waves (UGWs) and piezoelectric sensors offers a promising solution for monitoring aircraft structures. However, the challenge lies in the UGW signals generated by external impacts on plate structures, characterized by significant background noise, wide frequency bandwidth, multimodal interference, and inherent frequency dispersion. These complexities hinder the direct extraction of characteristic signals for impact localization in both time and frequency domains. Impact-induced UGW signals typically exhibit Gaussian amplitude-modulated and frequency-modulated (AM–FM) characteristics. Addressing this challenge, this article introduces the Gaussian reconstructed symplectic geometry mode decomposition (GRSGMD) method. GRSGMD combines power spectral density analysis and the Gaussian mixed-model clustering algorithm, to improve the discriminant function of the symmetric geometry component (SGC). This study investigates the decomposition performance and noise robustness of GRSGMD through simulated impact-induced UGW signals. Based on the adaptively extracted dominant energy narrowband of impact-induced UGW using GRSGMD, an impact source localization method is proposed, and verification experiments on aluminum plate impact localization are conducted. The results show that the GRSGMD method can effectively and adaptively extract the dominant energy narrowband of impact-induced UGW. The dominant energy narrowband of low-speed impact-induced UGW corresponds to the ${A}_{{0}}$ mode Lamb wave. The experimental results for impact source localization using GRSGMD align well with actual positions. The research findings contribute to providing an efficient and feasible signal processing method for impact monitoring based on UGW.

The inner curve radial piston motor plunger is easy to wear under radial force. The accuracy of fault identification for wear using a single signal is too low in the influence of fluid flow pulsations. In this article, sound and vibration signals are combined, and the required eigenvalues of the fused signals are extracted based on the feature importance index. The extracted feature values are fused using the Concat method to construct sound and vibration fusion vectors, which are combined with a Light Gradient Boosting Machine (LightGBM) model for training and diagnosis. A hydraulic motor plunger wears fault diagnosis method based on the fusion of vibration and sound signal features recognized by LightGBM is proposed. This method was utilized to analyze simulated plunger wear test data from an inner curve radial piston hydraulic motor test bed. The results show that the diagnostic accuracy using the LightGBM algorithm is improved by 1.35%, 2.21%, and 6.92% over AdaBoost, GBDT, and XGBoost algorithms under the diagnosis of acoustic-vibrational fusion signals, which are affected by the pulsation of liquid flow. The fault diagnosis method based on the fused signals recognized by LightGBM improved the diagnostic accuracy by 4.86% and 14.59%, respectively, compared to the diagnostic results using vibration or sound signals.

Aiming at the strong vibration condition of the inner curve radial plunger hydraulic motor, the bolt connection of the motor base is easy to loosen, and it is difficult to distinguish the early loosening faults online in time by using the vibration signals. In this article, we propose a bolt loosening fault diagnosis method based on LightGBM to recognize sound signal features, and this method can achieve online monitoring of bolt loosening faults. Through the vibration energy recovery test platform, the sound signals of four different bolt preload forces during the normal operation of the equipment were collected, and the bolt preload force was increased from completely loosened 0–60 $\text{N}\cdot \text{m}$ , with an increment of 20 $\text{N}\cdot \text{m}$ each time, and the sound signals were denoised using wavelet threshold denoising method. The LightGBM bolt loosening fault diagnosis model is constructed based on the gradient of one-sided sampling and mutually exclusive feature bundle algorithms. By extracting the time- and frequency-domain features of the denoised sound signals, a dataset containing labels of normal and three faulty signals is generated for training and diagnosis. Finally, the diagnostic accuracy of this method is compared and verified. The results show that the LightGBM algorithm after wavelet threshold denoising improves the diagnostic accuracy by 2.17% over the no-denoising LightGBM and by 5.47% and 2.21% over the XGboost algorithm before and after denoising, respectively.

Supervisory control and data acquisition (SCADA) is widely used in wind farms as an effective data acquisition system for wind turbines (WTs). However, in practical engineering applications, it is difficult for us to have adequate conditions to collect enough WT blade icing data, which leads to data imbalance and uneven distribution in the feature space. Using the classical synthetic minority oversampling technique (SMOTE) to balance the data may increase the overlap of positive and negative samples, or produce some redundant samples without useful information. A center jumping boosting machine (CJBM) method is proposed that combines an improved clustering-based oversampling (γ mini density peaks clustering SMOTE, γMiniDPC-SMOTE) and light gradient boosting machine (LightGBM) for blade icing prediction. First, to solve the problem of imbalanced and uneven distribution of WT data, a ${\gamma }$ MiniDPC-SMOTE method is proposed, which divides icing samples into multiple clusters, then increases icing samples, and alleviates uneven distribution in feature space. Second, calculating the intercept distance ${d}_{c}$ based on the binary search method and the adaptive selection of DPC parameters based on the step phenomenon of $\gamma $ parameters and verified by $\gamma $ -step of two WT icing data are proposed. Then, for the problem of low operating efficiency of the model under a large amount of imbalanced data, LightGBM is used for model training and icing prediction. Finally, validation was performed on two SCADA datasets. The results showed that the accuracy, precision, recall, F1-measure, and running times increased significantly, proving the superiority of the CJBM.

The liver is essential for animals and humans. Because of their low side effects and high safety, natural products have recently become a research hotspot for human health-related issues that can damage the liver. In this study, we investigated the protective effects in rats of Lycium barbarumbetaine (LBB) and Lycium barbarumbetaine Effervescent Tablet (LBBET) against liver injury caused by carbon tetrachloride (CCl4). The results showed that LBB and LBBET pretreatment significantly reduced the serum levels of alanine aminotransferase, aspartate transaminase (AST), and alkaline phosphatase, as well as the liver tissue levels of malondialdehyde. Meanwhile, glutathione peroxidase, and superoxide dismutase levels were significantly increased in liver tissues. In addition, LBB and LBBET may effectively alleviate CCl4-induced liver injury by a mechanism related to the activation of the Nrf2 signaling pathway. In conclusion, LBB and LBBET may serve as potential mitigators of CCl4-induced liver injury. Effervescent Tablet can be used as either a new formulation or practical product for patients who have difficulty swallowing regular tablets or capsules. This study provides a basis and new ideas for the development of functional foods or drugs related to the field of liver protection.