Journal of Vibroengineering

Journal of Vibroengineering, (in Press).Yuanhui Liang, Yuyu Zhu, Lei WangAutonomous underwater vehicles (AUVs) are indispensable equipments in underwater detection, surveying, and investigation. As the main power source of AUV, the accurate and timely fault diagnosis of thruster plays key role in ensuring its safe navigation. However, this task is full of challenges due to the complication, vagueness, and randomness of underwater settings. To address the issues, a hybrid diagnosis model named CLSTM-CNN-Attention is proposed, which combines convolutional long short term memory (CLSTM), convolutional neural network (CNN) and attention mechanism. Specifically, it combines an improved hybrid network to realize the connection of long-term and short-term memory (LSTM) with CNN in parallel, which can capture the time-related and space-related fault information of input signal simultaneously. At the same time, a new linear rectification function is also introduced into the hybrid model to enhance its anti-interference capability. Finally, the diagnostic performance of the hybrid model is further improved by adding attention mechanisms, which could better focus on the fused information. Experimental and comparison results indicate that the suggested approach has remarkable interference suppression capacity and surpasses other relevant methods, demonstrating good performance in fault diagnosis of AUV thruster.

Journal of Vibroengineering, (in Press).Huajie Zou, Fuhai CaiTo address the inadequacy of existing empirical formulas for estimating the vibration frequency of small airflow-induced piezoelectric generators, this study proposes a novel method for vibration frequency estimation specifically tailored to such generators. Through experimental analysis of vibration frequencies, the current expression for vibration frequency is modified to enhance the accuracy of frequency estimation for short resonators. The results indicate that the length of the resonator is the primary influencing factor. As the length increases, the frequency decreases, demonstrating an inverse relationship. The distance between the nozzle and the resonator is regarded as a secondary factor. Although there is a slight decrease in frequency as the spacing increases, its impact remains limited. Furthermore, the discrepancy between the theoretical values derived from the modified empirical frequency formula and the experimental data does not exceed 4 %, and the coefficient of determination (R2) for the modified empirical formulation is 0.9987, highlighting its reliability as an effective method for estimating vibration frequency. These conclusions may offer guidance in designing the vibration frequency and identifying essential structural parameters.

Journal of Vibroengineering, (in Press).Xiaolin Gao, Xi ChenIn response to the growing demand for advanced vibration and noise suppression in high-end equipment within the manufacturing sector, this paper proposes a novel design method for mechanical waveguide filters based on transmission line impedance matching theory. By establishing a precise force-electric analogy model, this approach effectively bridges mature circuit theory with mechanical dynamics, enabling the accurate prediction and control of elastic wave propagation. To address the prevalent issue of non-stationary vibration signals in industrial applications the Fractional Fourier Transform is innovatively employed for superior time-frequency analysis. The designed dual-joint matching filter validates this methodology, exhibiting outstanding performance at a center frequency of 2500 Hz with an insertion loss of 0.8 dB and a return loss of 19.1 dB. The filter also demonstrates sharp frequency selectivity, effective stopband suppression, and stable phase response, which are critical for maintaining signal integrity in precision systems. This research, which integrates theoretical modeling, equivalent circuit analysis, and advanced signal processing, provides a robust and efficient design framework. The proposed technique offers significant practical value for the mechanical industry, presenting a viable solution to enhance the dynamic performance, operational reliability, and noise control in modern mechanical systems, with direct applicability in fields such as aerospace, precision manufacturing, and automotive engineering.

Journal of Vibroengineering, (in Press).Jiyan Zeng, Yaohua Tong, Yujie Cheng, Chen LuAccurate prediction of the remaining useful life (RUL) of aircraft engines is crucial for ensuring flight safety and optimizing maintenance strategies. However, traditional data-driven methods typically yield point estimation, failing to quantify uncertainties arising from data noise and model limitations. This study proposes an attention-based ensemble method with partial-transfer Bayesian deep learning (Att-ensembled PT-BDL) for uncertainty quantification in aircraft engine RUL prediction. The proposed method transfers weights and biases from existing point estimation deep learning models as prior knowledge to the mean values of weights and biases in Bayesian deep learning models, freezing these parameters during training to reduce trainable parameters and enhance computational efficiency. An ensemble framework, enhanced by an attention mechanism, integrates multiple models to improve prediction accuracy and uncertainty quantification performance. A case study is conducted to demonstrate the effectiveness of the proposed method with a dataset of the PHM data challenge. The experiment results show that the proposed Att-ensembled PT-BDL method can achieve a better prediction accuracy and uncertainty quantification performance in terms of root mean square error (RMSE), prediction interval coverage probability (PICP) and prediction interval normalized average width (PINAW).

Journal of Vibroengineering, (in Press).Xufei Wang, Zhiqiang Xie, Ningchao Zhang, Jinde SongThe membrane air spring possesses the advantage of variable vertical stiffness. However, under limited vertical installation space, further investigation is required to analyze how variations in the rubber airbag cord parameters influence the vertical stiffness of membrane air springs. A finite element model of a specific membrane air spring was developed. Under a compression displacement of 40 mm, the effects of parameters such as the number of rubber airbag cord layers, cord spacing, cord angle, and cord cross-sectional area on vertical stiffness were examined. The results indicate that vertical stiffness decreases with an increase in cord spacing or cord angle, but increases with greater cord cross-sectional area or number of cord layers. Subsequently, the orthogonal experimental method was applied to construct an L9 (34) orthogonal table, and the influence of the four cord parameters on the vertical stiffness of the membrane air spring under 40 mm compression displacement was analyzed. The findings revealed that vertical stiffness is least affected by cord angle and most affected by the number of cord layers. The variation trend of vertical stiffness was determined by fitting the optimal parameter combination. Finally, the combination of four cord parameters yielding the maximum vertical stiffness was evaluated when the compression displacement increased to 50 mm, providing a reference for air spring configuration design in confined spaces.