Article
Automation & Control Systems
Xin Chen, Yu Guo, Jing Na
Summary: Synchronous averaging (SA) is a powerful signal processing tool that enhances the features of periodic events by suppressing nonsynchronous components. However, under random slip conditions, SA may not effectively enhance the features related to rolling element bearing (REB) faults. This article proposes two frameworks based on instantaneous angular speed (IAS) for synchronous averaging and introduces an improved negentropy indicator to characterize the richness of REB fault information. The effects of encoder resolution and structure damping factor on the waveform related to faulty REB are also studied. Simulation and experiment results demonstrate the effectiveness of the proposed schemes in enhancing the features of REB faults under random slip conditions.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Engineering, Mechanical
Dikang Peng, Yuejian Chen, Ming J. Zuo, Chris K. Mechefske
Summary: Accurately estimating instantaneous angular speed (IAS) is crucial for effective fault diagnosis under nonstationary operating conditions. This study proposes a method to enhance the accuracy of IAS obtained from different techniques and discusses three indicators that can assess the accuracy of the estimated IAS without a tachometer.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Mechanical
R. Bertoni, H. Andre
Summary: Early detection of bearing faults has been a major topic of research in the industry. Accelerometry is the most widely used method, but it has limitations in cases of high background noise or when environmental conditions do not allow proximity to the bearing. This paper proposes a fault detection method using incremental encoders to detect early bearing faults in a helicopter drive train.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Chemistry, Physical
Zirui Zhao, Xiaoli Wang, Yanqiang Hu, Zhihao Li, Lizhou Li, Liyan Wu
Summary: In this study, a wireless grease-lubricated triboelectric instantaneous angular speed sensor (GL-TEIASS) was proposed for bearing fault diagnosis. The sensor integrates a signal processing circuit and monitors the periodic fluctuation of instantaneous angular speed to diagnose bearing faults. Experimental results show that compared to dry friction, using grease reduces material mass loss and increases the electrical output. The sensor has a higher signal-to-noise ratio compared to traditional vibration signals, making it a promising technology for smart bearings.
Article
Engineering, Mechanical
Toby Verwimp, Alexandre Mauricio, Konstantinos Gryllias
Summary: This article investigates the viability of using a smartphone as a cost-effective tool for rotational speed extraction. The proposed method utilizes the rolling shutter camera of a smartphone to measure varying speeds, overcoming challenges such as changing lighting conditions and misalignment between the shaft and the camera.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Computer Science, Information Systems
Wei-Tao Zhang, Xiao-Fan Ji, Ju Huang, Shun-Tian Lou
Summary: The article presents a method based on the CCA criterion for blind extraction of specific fault signals from multi-channel observations, applicable to compound fault diagnosis of aero-engine spindle bearing. It utilizes the different fault characteristic frequencies of rolling element bearings to estimate delay parameters and optimizes the CCA criterion using the conjugate gradient method to improve reliability. Both simulated and experimental data are used to verify the algorithm's effectiveness in compound fault diagnosis.
Article
Engineering, Electrical & Electronic
Xiaolin Liu, Jiani Lu, Zhuo Li
Summary: In this paper, a fault diagnosis method based on multiscale fusion attention CNN (MSFACNN) is proposed to address the nonstationary characteristics of aircraft engine rolling bearings. By converting vibration signals into images and using multiscale convolution and attention mechanisms to extract and weigh fault features, higher recognition accuracy is achieved.
IEEE SENSORS JOURNAL
(2023)
Article
Acoustics
Baojia Chen, Zhichao Hai, Xueliang Chen, Fafa Chen, Wenrong Xiao, Nengqi Xiao, Wenlong Fu, Qiang Liu, Zhuxin Tian, Gongfa Li
Summary: This paper proposes a time-varying instantaneous frequency fault feature extraction method for rolling bearings under variable speed, which combines the improved multisynchrosqueezing transform, empirical Fourier decomposition, and generalized demodulation. The method can accurately extract fault features of rolling bearings and identify fault types.
JOURNAL OF SOUND AND VIBRATION
(2023)
Article
Engineering, Mechanical
J-h. Cai, Y-l. Xiao, L-y. Fu
Summary: By combining instantaneous spectrum estimation with FRFT and determining the optimal order based on the principle of maximum kurtosis coefficient, this method achieves more accurate characteristic frequency identification, providing a new approach for fault diagnosis of rolling bearing.
EXPERIMENTAL TECHNIQUES
(2022)
Article
Engineering, Electrical & Electronic
Jialiang Cui, Qianwen Zhong, Shubin Zheng, Lele Peng, Jing Wen
Summary: This study proposes a lightweight rolling bearing fault diagnosis method based on Gramian angular field and coordinated attention to improve recognition performance and diagnosis efficiency.
Article
Engineering, Electrical & Electronic
Yongbo Li, Ziwen Guo, Zhixiong Li, Zichen Deng, Khandaker Noman
Summary: Due to the coupling of multiple cylinders and high environmental noise, existing waveform analysis is inefficient in identifying the working status of large multicylinder marine diesel engines (MCMDEs). To solve this problem, a new framework called intrinsic multiscale dispersion entropy (IMDE) is proposed by calculating the MDE of intrinsically reconstructed instantaneous angular speed (IAS) signal. The IMDE can accurately extract fault features under different working conditions and outperforms existing techniques: MDE, MSE, and MFE.
IEEE SENSORS JOURNAL
(2023)
Article
Automation & Control Systems
Junqiang Liu
Summary: Gas-path fault diagnosis of an aero-engine is a key challenge for flight safety. This study proposes a new integrative diagnostic approach using the HELM-TL model and attention mechanism to improve diagnosis performance and address the over-fitting problem. Experimental results confirm the effectiveness of the proposed method.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Mathematics
Zitong Yan, Hongmei Liu
Summary: A new unsupervised representation learning method called Signal Momentum Contrast (SMoCo) is proposed in this study, which extracts fault features using the contrastive learning algorithm - Momentum Contrast (MoCo), and performs feature extraction using the data augmentation method.
Article
Engineering, Electrical & Electronic
Juanjuan Shi, Zehui Hua, Weiguo Huang, Patrick Dumond, Changqing Shen, Zhongkui Zhu
Summary: This article proposes a new method for time-frequency analysis of vibration signals for bearing fault diagnosis. The method uses instantaneous frequency synchronized generalized stepwise demodulation transform, which overcomes the limitations of traditional methods in analyzing bearing vibrations. The effectiveness of the proposed method is validated through simulations and measured data, and comparisons with other methods demonstrate its superiority.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Mechanical
Adrien Marsick, Hugo Andre, Ilyes Khelf, Quentin Leclere, Jerome Antoni
Summary: This paper investigates the restoration of cyclostationarity of rolling element bearing signals, which are known to exhibit pseudo-cyclostationary properties, limiting the efficiency of health monitoring. The paper presents a synthesis of the state of art on this issue, introducing the concept of cycle of reference (CoR) for cyclostationarity. It explores the limitations of the cyclostationary theoretical framework and introduces the use of shaft cycle of reference to restore cyclostationarity, along with a method to estimate the appropriate angle-time relationship. The paper also explains demodulation parameters, the introduction of unwanted artefacts, and illustrates the superiority of cyclostationarity restoration for monitoring purposes through two industrial cases.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Automation & Control Systems
Hao Fu, Lei Deng, Baoping Tang, Chunhua Zhao, Yi Huang
Summary: This article proposes a time-space incentives control algorithm to dynamically improve the topology equilibrium for wireless sensor networks used in mechanical vibration monitoring systems. Multiple relevant parameters that influence the network topology are designed and adopted, and two calculation methods are devised to determine the parameter weights. A static evaluation model is constructed to measure the performance of the network topology in the space dimension. Based on this model, the proposed algorithm dynamically adjusts the network topology by changing the transmission power of each node considering the comprehensive evaluation value.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Automation & Control Systems
Qichao Yang, Baoping Tang, Qikang Li, Xiaoli Liu, Lei Bao
Summary: This paper proposes a novel prediction framework for forecasting the remaining useful life (RUL) of aircraft engines. The framework utilizes a dual-frequency enhanced attention network architecture, separable convolutional neural networks, frequency-enhanced modules, and an efficient channel attention block to improve the prediction performance and robustness of the model.
Article
Physics, Applied
Yi Huang, Guobiao Hu, Chaoyang Zhao, Baoping Tang, Xiaojing Mu, Yaowen Yang
Summary: A novel L-shaped self-adaptive piezoelectric energy harvester (LSA-PEH) with a slider is proposed in this study for harvesting vibration energy. A linearized mathematical model is established to predict the resonant frequency of the LSA-PEH based on the position of the slider. Experimental results show that the slider of the proposed LSA-PEH can passively relocate its position to adjust its resonant frequency and maintain resonance. The proposed LSA-PEH has a 350% increased bandwidth compared to a conventional L-shaped beam harvester.
JOURNAL OF PHYSICS D-APPLIED PHYSICS
(2023)
Article
Engineering, Multidisciplinary
Cheng Li, Yi Wang, Guangyao Zhang, Yi Qin, Baoping Tang
Summary: This paper presents an enhanced approach for tacholess order tracking in wind turbine gearboxes, which improves the accuracy and efficiency of instantaneous angular speed estimation. The approach combines a Vold-Kalman filter, Hilbert transform, and resampling technique to extract the instantaneous phase from vibration signals and calculate the envelope order spectrum for fault diagnosis. Experimental results show that the presented approach outperforms state-of-the-art methods in terms of accuracy and efficiency.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Liaoyuan Huan, Baoping Tang, Chunhua Zhao
Summary: To address the problem of limited storage and computing resources in wireless sensor networks (WSNs), a global composite compression method for deep neural networks (DNN) is proposed. The method removes redundant parameters and kernels through coarse and fine-grained composite pruning, and further reduces model storage and improves inference speed through quantification of output features and weight parameters. Experimental results show that the proposed method achieves a compression rate of approximately 20x, maintains high diagnostic accuracy, reduces power consumption, and improves system time, indicating advanced performance in DNN model compression, node power consumption, and data transmission delay.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Kai Zhang, Zhixuan Li, Qing Zheng, Guofu Ding, Baoping Tang, Minghang Zhao
Summary: This study presents a bidirectional guidance method that enables deep networks to diagnose faults robustly with the ability to label noise tolerance. The proposed method uses discrete wavelet packet transform (DWPT) to preprocess vibration signals and a bidirectional loss (BL) function is used to improve the robustness of the model in the case of noisy labels. A fast cosine decay (FCD) strategy of the learning rate is also designed to further boost the recognition performance of the model under severe noisy labels. The experimental results demonstrate that the proposed method outperforms other cutting-edge methods in fault diagnosis with severely noisy labels.
IEEE SENSORS JOURNAL
(2023)
Article
Automation & Control Systems
Chunhua Zhao, Baoping Tang, Lei Deng, Yi Huang, Qikang Li
Summary: This article proposes a novel multilevel adaptive near-lossless compression method in edge collaborative WSN for mechanical vibration monitoring, addressing the challenges of limited storage and computational resources as well as high delay in transmitting massive vibration data. The proposed method characterizes mechanical fault feature information accurately in low storage space and integrates edge computing technique to reduce the pressure on data center servers. Comprehensive experiments demonstrate that the proposed approach achieves high-precision data reconstruction and feature detection, significantly improving the computational power and transmission efficiency of WSN.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Automation & Control Systems
Chunhua Zhao, Baoping Tang, Yi Huang, Lei Deng
Summary: This article proposes a novel edge collaborative compressed sensing approach for mechanical vibration monitoring. It combines compressed sensing and edge computing to address the issues of limited storage, computational resources, and delay in transmitting vibration data in wireless sensor networks. The approach enhances the acquisition efficiency and computing capacity of the network and provides a fast convergence convex optimization algorithm for data reconstruction. Comprehensive experiments demonstrate that the proposed method reduces transmission data by 70% while maintaining high-precision fault detection and improving acquisition and transmission efficiencies. It offers a potential solution for practical engineering applications.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Quan Qian, Yi Wang, Taisheng Zhang, Yi Qin
Summary: The transfer diagnosis performance of deep domain adaptation methods is completely determined by the discrepancy representation metric. The commonly used metric, maximum mean discrepancy (MMD), based on the mean statistic, has poor discrepancy representation in some cases. In this study, the authors theoretically explore the relationship between MMD and kernel function and propose a novel discrepancy representation metric called maximum mean square discrepancy (MMSD). The experimental results show that MMSD has a better ability of discrepancy representation and a higher diagnosis accuracy compared with other well-known metrics.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Multidisciplinary
Chengyuan Chen, Yi Wang, Hulin Ruan, Yi Qin, Baoping Tang
Summary: In this article, a robust intelligent fault diagnosis approach is proposed, which uses deep neural networks and a mixture model to achieve accurate fault diagnosis under noisy labels.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2023)
Article
Engineering, Mechanical
Qichao Yang, Baoping Tang, Shilong Yang, Yizhe Shen
Summary: This paper proposes a network framework called DR-DTPN, which integrates data repair and degradation trend prediction to address the serious deviation in equipment degradation trend prediction caused by missing monitoring data and distribution changes. DR-DTPN considers the trend and periodic variations of the signal and dynamically infers the latent vector of the spatial spectrum. By encoding the shared temporal dynamics of the training window as polynomial and trigonometric functions, and inferring the spatial spectral latent vector through convex optimization, DR-DTPN captures the current temporal dynamics and correlations between features. DR-DTPN simultaneously interpolates past missing values and predicts future degradation trend values through optimal latent space spectral decomposition. The data interpolation and prediction performance of DR-DTPN have been validated on the PHM2012 bearing degradation data and further verified through engineering applications on wind turbines and aero engines.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Instruments & Instrumentation
Mingxiang Ling, Shilei Wu, Zhihong Luo, Liguo Chen, Tao Huang
Summary: This paper proposes a new electro-mechanical dynamic stiffness matrix of piezoelectric stacks for systematic analysis, which enables the full acquisition of parameters such as the time-domain response of piezoelectric hysteresis and the dynamic resonance behavior of compliant mechanisms.
SMART MATERIALS AND STRUCTURES
(2023)
Article
Engineering, Electrical & Electronic
Xiaomeng Li, Yi Wang, Hulin Ruan, Baoping Tang, Yi Qin
Summary: This article proposes a weak signal detection method based on deep manifold learning (DML). The method constructs a mapping model between the noisy waveform feature manifold (WFM) and the pure repetitive impulses, and adaptively mines the high-dimensional WFM through deep compression. The mined signal is considered as the detected signal.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Electrical & Electronic
Xin Zhang, Wensheng Hou, Yi Wang, Ning Jiang
Summary: Based on action observation and brain-computer interface technology, this study successfully achieved lower latency and higher accuracy by detecting motion onset visual evoked potentials to distinguish intentional and nonintentional control states. The analysis of EEG data from seniors revealed that the designed AO stimulus can elicit mVEP, which was detected using the $P+L$ method with the highest accuracy and no significant difference between age groups. These findings contribute to the improvement of BCI technology and stroke rehabilitation.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Acoustics
Xiaomeng Li, Yi Wang, Xiang Wan, Baoping Tang, Yi Qin, Caibin Xu
Summary: A physics-informed deep filtering method is proposed to enhance ultrasonic guided waves for early-stage defect inspection. The method constructs and trains a deep learning model based on template signals of ultrasonic guided waves, and then applies it to enhance weak ultrasonic guided waves from practical cases. Experimental results show that the proposed method outperforms traditional approaches.
JOURNAL OF SOUND AND VIBRATION
(2023)