Article
Engineering, Electrical & Electronic
Zhiliang Liu, Huan Wang, Junjie Liu, Yong Qin, Dandan Peng
Summary: This article explores the potential of using multitask learning to improve the fault diagnosis performance of bearings, introducing speed identification task and load identification task as auxiliary tasks, and proposing a multitask one-dimensional convolutional neural network (MT-1DCNN). Experimental results demonstrate that multitask learning can enhance the fault diagnosis performance of the network.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Chemistry, Multidisciplinary
Cheng-Jian Lin, Chun-Hui Lin, Frank Lin
Summary: A vector-based convolutional fuzzy neural network (vector-CFNN) was developed in this study to diagnose faults from signals. The fusion layer played a crucial role in combining characteristics and improving the model's performance. Experimental results showed that vector-CFNN outperformed other neural networks in terms of accuracy and parameter efficiency, making it feasible for online spindle vibration monitoring.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Multidisciplinary
Huichao Zhang, Peiming Shi, Dongying Han, Linjie Jia
Summary: Due to the uncertainty of the industrial environment, the effective features of multivariate data are obscured by environmental noise. The limited ability of single signal analysis method to recognize these features necessitates the proposed method of adaptive multivariate variational mode decomposition and multi-scale convolutional neural network for rolling bearing fault diagnosis. Experimental results using bearing data from Paderborn University show that the proposed method achieves a fault diagnosis accuracy of 98.60% under the same conditions, indicating its practical significance and advantages.
Article
Energy & Fuels
Ziran Guo, Ming Yang, Xu Huang
Summary: This paper proposes a bearing fault diagnosis method based on the motor speed signal. The feature extraction and analysis of the rotational speed signal are carried out using CNN, and an improved algorithm is proposed. The experimental results show that this method can effectively diagnose bearing faults even in the presence of misalignment fault interference.
Article
Engineering, Multidisciplinary
Jing Zhao, Shaopu Yang, Qiang Li, Yongqiang Liu, Xiaohui Gu, Wenpeng Liu
Summary: In this paper, a fault feature extraction method based on deep learning is proposed, utilizing data augmentation and signal-to-image mapping techniques. A convolutional neural network model is established to extract fault features and achieve fault classification.
Article
Chemistry, Analytical
Yuanqing Luo, Wenxia Lu, Shuang Kang, Xueyong Tian, Xiaoqi Kang, Feng Sun
Summary: In this study, an enhanced feature extraction network (EFEN) based on acoustic signal feature learning is proposed for fault diagnosis of rolling bearings. Experimental results show that the EFEN network achieves high accuracy in fault diagnosis and effectively detects rolling bearing faults based on acoustic signals.
Article
Engineering, Electrical & Electronic
Zonghao Yuan, Zengqiang Ma, Xin Li, Jiajing Li
Summary: To address the complex and noisy working condition of high-speed train wheelset bearings, a graph convolution network (GCN) model is proposed. A weighted e-recurrence network is used to construct a recurrence graph (RG) to analyze the recurrence relationship between samples, and a maximum mean discrepancy (MMD) is used to reduce the weight gap between different recursive pairs. A vertex mean normalization (MN) layer is proposed to normalize individual vertex features and improve training stability. A multichannel MN-GCN is also proposed to fit multiscale graph features and improve antinoise ability. Experimental results show improved fault diagnosis performance of the proposed model.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Yongjian Sun, Shaohui Li
Summary: This article proposes a novel fault diagnosis method based on convolutional neural networks (CNN), which does not require human intervention and achieves high accuracy. By transforming vibration signals into symmetrical images and inputting them into the CNN, the fault type can be automatically diagnosed.
Article
Engineering, Electrical & Electronic
Jianbo Yu, Xingkang Zhou, Liang Lu, Zhihong Zhao
Summary: This article proposes a new CNN, multiscale fusion global sparse network (MFGSNet) for feature extraction from vibration signals and gearbox fault diagnosis. It outperforms typical DNNs like ResNet and DenseNet in gearbox fault diagnosis.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Artificial Intelligence
Xiaohan Chen, Beike Zhang, Dong Gao
Summary: This study introduces an automatic feature learning neural network based on raw vibration signals, which utilizes convolutional neural networks and long short-term memory to extract signal characteristics, achieving an accuracy of 98.46%.
JOURNAL OF INTELLIGENT MANUFACTURING
(2021)
Article
Engineering, Electrical & Electronic
Linshan Jia, Tommy W. S. Chow, Yu Wang, Yixuan Yuan
Summary: A novel fault diagnosis framework called MRA-CNN is proposed in this article to learn discriminative multiscale features from vibrational signals and reduce noises. Experimental results show that the proposed method achieves higher accuracy in highly noisy environments compared to state-of-the-art methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Electrical & Electronic
Ruixian Li, Li Zhuang, Yongxiang Li, Changqing Shen
Summary: This study develops an intelligent data-driven method for bearing fault diagnosis, which recognizes the health condition of bearings through feature transformation and fault recognition based on transformed features. The method demonstrates effectiveness and robustness in noisy environments, showing promising performance on datasets with higher noise levels.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Shuzhi Gao, Zhiming Pei, Yimin Zhang, Tianchi Li
Summary: This paper proposes an adaptive convolutional neural network based on Nesterov momentum for rolling bearing fault diagnosis, which improves both the accuracy and convergence of neural networks compared to traditional methods by replacing the traditional momentum and utilizing an adaptive learning rate rule to enhance the generalization ability of the network.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Multidisciplinary
Huijie Ma, Shunming Li, Jiantao Lu, Siqi Gong, Tianyi Yu
Summary: This paper proposes an improved fault feature extraction method called KurHLSC based on probability sparse coding, which can accurately extract weak bearing fault features even in strong noise.
Article
Engineering, Multidisciplinary
Chunran Huo, Weiyang Xu, Quansheng Jiang, Yehu Shen, Qixin Zhu, Qingkui Zhang
Summary: This article proposes a dual sample screening method based on predicted label changes to improve the target domain prediction ability in fault diagnosis of rolling bearings. By eliminating prediction errors of pseudo-labels during training, the method enhances the stability and diagnostic accuracy of the network.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Yitian Wang, Yuxiang Wei, Huan Wang
Summary: This study introduces a class imbalanced wafer defect classification method based on Variational Autoencoder Generative Adversarial Network (VAE-GAN), aiming to solve the data imbalance problem in wafer defect classification. The VAE-GAN is responsible for generating new samples to address the imbalance, while the classifier is used for classifying wafer defect patterns. The experimental results demonstrate that the samples generated by VAE-GAN significantly improve the performance of the wafer defect classification system.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2023)
Article
Transportation
Jing Shi, Muhammad Hussain, Dandan Peng
Summary: This study aims to analyze the factors contributing to aberrant driving behaviors and road accidents among Chinese ride-hailing drivers. The findings show that traditional taxi drivers are more prone to aberrant driving behaviors compared to private car drivers. Male and young ride-hailing drivers are more likely to engage in risky violations. Additionally, risky violations and work-condition factors have an impact on road accidents. These findings provide insights into the driving characteristics of ride-hailing drivers and can contribute to the development of more effective policies to reduce road accidents caused by them.
JOURNAL OF TRANSPORTATION SAFETY & SECURITY
(2023)
Article
Engineering, Electrical & Electronic
Yuhua Yin, Zhiliang Liu, Mingjian Zuo, Zetong Zhou, Junhao Zhang
Summary: In this article, a three-dimensional data compression method based on fault mechanism is proposed, which effectively reduces the data size by introducing the concept of compression dimension and bit cost, while maintaining good diagnostic performance.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Artificial Intelligence
Chuanjiang Li, Shaobo Li, Huan Wang, Fengshou Gu, Andrew D. Ball
Summary: Deep learning-based fault diagnosis methods have achieved remarkable progress, but they are often coarse grained and require large amounts of data, which cannot identify the root causes of mechanical system failures at a finer granularity with limited fault data. In this study, a novel attention-based deep meta-transfer learning (ADMTL) method is proposed to address the challenges of fine-grained fault feature extraction and limited model generalization ability. The proposed method achieves excellent performance in few-shot fine-grained fault diagnosis tasks.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Chemistry, Analytical
Yuhua Yin, Zhiliang Liu, Junhao Zhang, Enrico Zio, Mingjian Zuo
Summary: In this paper, an adaptive sampling framework of segment intervals is proposed to monitor mechanical degradation. The results of the experiments show that the proposed method has better sampling effects compared to existing methods, and the results are closely related to data status and degradation indicators.
Article
Chemistry, Analytical
Zhe Xie, Guoli Zhu, Dailin Zhang, Dandan Peng, Jinlong Hu, Yueyu Sun
Summary: To achieve automatic disc cutter replacement of shield machines, measuring the accurate pose of the disc cutter holder by machine vision is crucial. However, under polluted and restricted illumination conditions, achieving pose estimation by vision is a great challenge. This paper proposes a line-features-based pose estimation method for the disc cutter holder of the shield machine by using a monocular camera. The experimental result shows that the proposed pose estimation method is highly reliable and can meet the measurement accuracy requirements in practical engineering applications.
Article
Automation & Control Systems
Yuxiang Wei, Huan Wang
Summary: This paper proposes a noise-robust framework for wafer defect recognition, which utilizes discrete wavelet transform for frequency learning. It introduces a learnable discrete wavelet transform layer and a frequency-location attention module. Experimental results show that the framework achieves excellent performance in detecting wafer defect images, with an accuracy of 98.84%, and outperforms other methods under high noise ratios.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Automation & Control Systems
Zhiliang Liu, Liyuan Ren
Summary: This study proposes a method based on morphological image processing to suppress shaking noise and improve the signal-to-noise ratio of magnetic flux leakage (MFL) signals, in order to monitor the health condition and detect flaws in steel wire rope (SWR).
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Automation & Control Systems
Le Sun, Rui Zhou, Dandan Peng, Athman Bouguettaya, Yanchun Zhang
Summary: Building a quality service-based system is an important research topic in software engineering. Existing keyword-based methods for building such systems do not allow relaxation of the function requirements. To address this, we propose a new problem and two algorithms to solve it.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Energy & Fuels
Yipu Wang, Huan Wang
Summary: Proposed a wavelet attention-powered hierarchical dynamic frequency learning framework (WAPHF) that integrates CNN and wavelet transform to learn frequency features in the frequency domain. The framework is able to dynamically select valuable frequency features using a dynamic frequency-focused attention (DFFA) module. Experimental results show that the framework accurately predicts the state of health of batteries and outperforms state-of-the-art methods.
JOURNAL OF ENERGY STORAGE
(2023)
Article
Health Care Sciences & Services
Shangzhi Xiong, Wei Jiang, Ruilin Meng, Chi Hu, Hui Liao, Yongchen Wang, Chang Cai, Xinyi Zhang, Pengpeng Ye, Yanqiuzi Ma, Tingzhuo Liu, Dandan Peng, Jiajuan Yang, Li Gong, Qiujun Wang, David Peiris, Limin Mao, Maoyi Tian
Summary: The study assessed China's primary health care (PHC) system to understand the factors influencing the uptake of the National Essential Public Health Service Package (NEPHSP) for hypertension and type-2 diabetes (T2DM) management. The results showed that despite the government's efforts to strengthen the PHC system, there are still many barriers, including insufficient and under-qualified PHC personnel, gaps in medicines and equipment, fragmented health information systems, low trust and utilization of PHC among residents, challenges in coordinated and continuous care, and lack of cross-sectorial collaborations.
LANCET REGIONAL HEALTH-WESTERN PACIFIC
(2023)
Article
Medicine, Research & Experimental
Dandan Peng, Tingmei Zhao, Weiqi Hong, Minyang Fu, Cai He, Li Chen, Wenyan Ren, Hong Lei, Jingyun Yang, Aqu Alu, Yanghong Ni, Jian Liu, Jiong Li, Wei Wang, Guobo Shen, Zhiwei Zhao, Li Yang, Jinliang Yang, Zhenling Wang, Yoshimasa Tanaka, Guangwen Lu, Xiangrong Song, Xiawei Wei
Summary: BA.4 and BA.5 (BA.4/5), the subvariants of Omicron, are more transmissible and have stronger immune evasion capability compared to BA.1. Hence, a third booster vaccination is urgently needed for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) due to these variants. Heterologous boosters, especially with a third heterologous protein subunit booster, may provide more effective immunity. In this study, a Delta full-length spike protein sequence-based mRNA vaccine was used as the priming shot, and a recombinant trimeric receptor-binding domain (RBD) protein vaccine (RBD-HR/trimer) was developed as a third heterologous booster. The heterologous group showed higher neutralizing antibody titers and stronger cellular immune response against BA.4/5-included SARS-CoV-2 variants compared to the homologous mRNA group. The RBD-HR/trimer vaccine is a suitable candidate for booster immune injection.
Article
Engineering, Electrical & Electronic
Wenjun Luo, Huan Wang
Summary: Wafer defect pattern recognition is crucial for optimizing the manufacturing process of semiconductor products. To tackle the challenge of recognizing multiple defect types on a single wafer map, we propose a composite wafer defect recognition framework (CWDR-Net) with a multiview dynamic feature enhancement module and a class-specific classifier. Our framework selectively extracts information from the defect pattern and class-specifically recognizes each basic defect type using a feature enhancement module that dynamically enhances the features from three different perspectives. Experimental results on a real dataset demonstrate the effectiveness of our framework in recognizing composite defects.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Electrical & Electronic
Shicheng Pei, Huan Wang, Te Han
Summary: This article introduces a time-efficient NAS-based AUV fault diagnosis framework (TENAS-FD), which can quickly search for excellent network architectures for AUV fault diagnosis. A novel scoring algorithm is constructed to evaluate the performance of untrained networks. Experimental results show that TENAS-FD has better diagnostic performance compared to hand-designing models.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Green & Sustainable Science & Technology
Huan Wang, Yan-Fu Li, Ying Zhang
Summary: This study proposes a method for battery health state monitoring based on spiking neural networks and electrochemical impedance spectroscopy, achieving accurate SOH estimation through simulating the feature processing mechanism of brain neurons and low power consumption.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2023)