Article
Engineering, Multidisciplinary
Zhi Zheng, Jiuman Fu, Chuanqi Lu, Yong Zhu
Summary: The article introduces a new transfer learning network for processing fault signals of rolling bearings, which has a novel structure and unique training strategy, achieving low computation cost, high accuracy, and strong diagnosis ability through a new optimization method.
Article
Engineering, Electrical & Electronic
Haopeng Liang, Jie Cao, Xiaoqiang Zhao
Summary: This study proposes a small sample rotating machinery fault diagnosis method based on a multibranch and multiscale dynamic convolutional network (MBSDCN), which can accurately diagnose faults under small sample and noisy conditions through the feature splitting strategy, channel reconstruction attention mechanism, and novel multiscale feature extraction model.
IEEE SENSORS JOURNAL
(2023)
Article
Automation & Control Systems
Yunjia Dong, Yuqing Li, Huailiang Zheng, Rixin Wang, Minqiang Xu
Summary: This paper proposes an intelligent fault diagnosis framework based on dynamic model and transfer learning for rolling element bearings. It addresses the small sample problem by generating simulation data using a dynamic model and applying the diagnosis knowledge gained from the simulation data to real scenarios through parameter transfer strategies, ultimately improving the fault identification performance.
Article
Automation & Control Systems
Jun-Wei Zhu, Bo Wang, Xin Wang
Summary: This paper proposes a fault diagnosis method for chemical processes with small samples by using data self-generation and transfer learning to expand fault samples and adopting a model-based transfer learning strategy to improve the robustness of the method. Additionally, a sample reconstruction-based convolutional neural network is introduced to adaptively extract features from both time and spatial domains, enabling the identification of fault types in industrial processes with small samples.
INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS
(2023)
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
Chunran Huo, Quansheng Jiang, Yehu Shen, Chenhui Qian, Qingkui Zhang
Summary: A modified ADC-CNN model and improved LATL training method were proposed for fault diagnosis of rolling bearings. Experimental results showed that the proposed method significantly outperformed the traditional 1D-CNN model on the PU dataset.
Article
Automation & Control Systems
Na Lu, Huiyang Hu, Tao Yin, Yaguo Lei, Shuhui Wang
Summary: A novel transfer relation network (TRN) has been developed, combining few-shot learning mechanism and transfer learning, to address multidomain problems in fault diagnosis. By utilizing Multikernel Maximum Mean Discrepancy (MK-MMD) and episode-based few-shot training strategy, efficient domain feature transfer and diagnosis have been achieved.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Automation & Control Systems
Chunran Huo, Quansheng Jiang, Yehu Shen, Qixin Zhu, Qingkui Zhang
Summary: Deep transfer learning is used to solve the problem of unsupervised intelligent fault diagnosis of rolling bearings. However, when the data distribution between two domains is different, the existing deep transfer learning models are not enough to complete the target domain data learning. An enhanced transfer learning method based on the linear superposition network is proposed for rolling bearing fault diagnosis. Experimental results show improved bearing fault diagnostic precision compared to traditional feature-based transfer learning methods.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Xinglong Pei, Xiaoyang Zheng, Jinliang Wu
Summary: The paper introduces a novel Transformer convolution network (TCN) based on transfer learning, which has achieved highly accurate fault diagnosis. Experimental results demonstrate the robustness and effectiveness of the proposed method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Artificial Intelligence
Supriya Asutkar, Chaitravi Chalke, Kajal Shivgan, Siddharth Tallur
Summary: TinyML, embedded with powerful machine learning algorithms in low-cost edge devices, has the potential to enable smart sensor nodes for fault diagnosis. This study presents a transfer learning framework with TinyML-powered convolutional neural networks (CNNs) for vibration-based fault diagnosis of different machines, achieving higher accuracy compared to conventional approaches. The use of transfer learning also allows for high accuracy with a limited number of training samples.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Anurag Choudhary, Tauheed Mian, Shahab Fatima, B. K. Panigrahi
Summary: This paper proposes an intelligent Passive Thermography (PTG) based fault diagnosis technique for detection of bearing faults using Convolutional Neural Network (CNN) with Transfer Learning (TL) under varying working conditions. The experimental results demonstrate that the suggested approach could successfully learn transferable characteristics from the source domain model and cope with the issue of limited availability of training data required for the target domain. The classification accuracy on the target domain dataset ranges from 89-95.4% (IM dataset) and 96.5-97.5% (MFS dataset), showing the benefits of the suggested method as an effective non-invasive diagnostic tool for rotating machines to avoid unexpected system shutdowns.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Shuwen Chen, Hongjuan Ge, Huang Li, Youchao Sun, Xiaoyan Qian
Summary: Convolutional neural networks (CNNs) can automatically extract features. This paper discusses the design process of the developed discrete time-series convolution neural network (DTCNN) and develops a hierarchical method for TRUs fault diagnosis, along with a transfer learning-based fault diagnosis method. The results show that transfer learning is an effective way to construct the diagnosis network for similar equipment and often leads to better performance.
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
Green & Sustainable Science & Technology
Yan Zhang, Wenyi Liu, Xin Wang, Heng Gu
Summary: This paper presents a fault diagnosis method for identifying different fault conditions in wind turbines' rolling bearings and gears. Compressed sensing technology is used for noise reduction and feature extraction. The fault diagnosis scheme combines deep transfer learning and convolutional neural network (DTL-CNN) for fault type identification. Experimental results demonstrate the reliability and superiority of the proposed method in wind turbine fault diagnosis of rolling bearings and gears.
Article
Computer Science, Information Systems
Haibin Huangfu, Yong Zhou, Jianxin Zhang, Shangjun Ma, Qian Fang, Ye Wang
Summary: In this paper, an inter-turn short-circuit fault in a permanent magnet synchronous motor was analyzed and a fault diagnosis method based on transfer learning with a VGG16 convolution network was proposed. The experimental results showed that the proposed method can effectively and accurately identify faults, and has good engineering guidance significance.
Article
Computer Science, Artificial Intelligence
Haotian Shi, Chengjin Qin, Dengyu Xiao, Liqun Zhao, Chengliang Liu
KNOWLEDGE-BASED SYSTEMS
(2020)
Article
Computer Science, Artificial Intelligence
Jianfeng Tao, Chengjin Qin, Dengyu Xiao, Haotian Shi, Xiao Ling, Bingchu Li, Chengliang Liu
JOURNAL OF INTELLIGENT MANUFACTURING
(2020)
Article
Engineering, Multidisciplinary
Chengjin Qin, Jianfeng Tao, Haotian Shi, Dengyu Xiao, Bingchu Li, Chengliang Liu
PRECISION ENGINEERING-JOURNAL OF THE INTERNATIONAL SOCIETIES FOR PRECISION ENGINEERING AND NANOTECHNOLOGY
(2020)
Article
Computer Science, Interdisciplinary Applications
Haotian Shi, Haoren Wang, Chengjin Qin, Liqun Zhao, Chengliang Liu
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2020)
Article
Computer Science, Artificial Intelligence
Dengyu Xiao, Chengjin Qin, Honggan Yu, Yixiang Huang, Chengliang Liu
Summary: The previous deep learning technology faces challenges in motor fault diagnosis due to the complexity and variability of working conditions, making it difficult to extract effective fault diagnostic representations.
JOURNAL OF INTELLIGENT MANUFACTURING
(2021)
Article
Computer Science, Artificial Intelligence
Yanrui Jin, Chengjin Qin, Yixiang Huang, Wenyi Zhao, Chengliang Liu
KNOWLEDGE-BASED SYSTEMS
(2020)
Article
Automation & Control Systems
Chengjin Qin, Jianfeng Tao, Haotian Shi, Dengyu Xiao, Bingchu Li, Chengliang Liu
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2020)
Article
Computer Science, Artificial Intelligence
Yanrui Jin, Chengjin Qin, Jinlei Liu, Ke Lin, Haotian Shi, Yixiang Huang, Chengliang Liu
KNOWLEDGE-BASED SYSTEMS
(2020)
Article
Engineering, Multidisciplinary
Yanrui Jin, Chengjin Qin, Yixiang Huang, Chengliang Liu
Summary: This paper introduces a novel deep learning method for compound fault diagnosis, achieving high accuracy through a decoupling attentional residual network, multi-label decoupling classifier, and active learning approach. The method can reach the same accuracy with a small number of compound fault samples as using a large number of samples, reducing the labeling workload for domain experts.
Article
Construction & Building Technology
Honggan Yu, Jianfeng Tao, Sheng Huang, Chengjin Qin, Dengyu Xiao, Chengliang Liu
Summary: This study proposes a method for estimating the wear of each disc cutter in real-time based on field parameters. By constructing a new health index and establishing a mapping model, the prediction accuracy is improved. Results from field data validation show that the method can accurately estimate cutter wear in real-time.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Engineering, Mechanical
Chengjin Qin, Gang Shi, Jianfeng Tao, Honggan Yu, Yanrui Jin, Junbo Lei, Chengliang Liu
Summary: This study introduces a novel hybrid deep neural network (HDNN) for accurately predicting cutterhead torque for shield tunneling machines, achieving high prediction accuracy under different geological conditions.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Chemistry, Analytical
Xiangpeng Liu, Danning Wang, Yani Li, Xiqiang Guan, Chengjin Qin
Summary: Advancements in deep learning and computer vision have led to the discovery of effective solutions for improving the precision of autonomous harvesting in the field of agricultural automation. In this article, the DA-Mask RCNN model is proposed, which utilizes depth information to improve detection accuracy. Experimental results demonstrate the effectiveness of the proposed model.
Article
Chemistry, Multidisciplinary
Rui Tang, Chengjin Qin, Mengmeng Zhao, Shuang Xu, Jianfeng Tao, Chengliang Liu
Summary: This paper proposes an optimized fractional-order PID (FOPID) method to suppress vibrations of high-speed elevators. The effectiveness of the controller in reducing elevator vibration was verified through numerical simulation. The results indicate that the FOPID controller significantly reduces horizontal acceleration compared to no controller and conventional PID controller.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Electrical & Electronic
Chao Liu, Chengjin Qin, Xi Shi, Zengwei Wang, Gang Zhang, Yunting Han
Summary: TScatNet is a cross-domain diagnosis model that extracts domain-invariant features using Morlet wavelet, modulus, and scaling averaging, performing well under various working conditions.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Proceedings Paper
Engineering, Multidisciplinary
Dengyu Xiao, Zhiyu Tao, Chengjin Qin, Honggan Yu, Yixiang Huang, Chengliang Liu
2020 PROGNOSTICS AND SYSTEM HEALTH MANAGEMENT CONFERENCE (PHM-BESANCON 2020)
(2020)