A5036102635- Machine Fault Diagnosis Techniques
- Fault Detection and Control Systems
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Magnetic properties of thin films
- Anomaly Detection Techniques and Applications
- Non-Destructive Testing Techniques
- Magnetic Field Sensors Techniques
- Industrial Vision Systems and Defect Detection
- Physics of Superconductivity and Magnetism
- Electrocatalysts for Energy Conversion
- Electrochemical Analysis and Applications
- Fuel Cells and Related Materials
- Advanced machining processes and optimization
- Structural Health Monitoring Techniques
- Magnetic and transport properties of perovskites and related materials
- Domain Adaptation and Few-Shot Learning
- Imbalanced Data Classification Techniques
- Mineral Processing and Grinding
- Virtual Reality Applications and Impacts
- Magneto-Optical Properties and Applications
- Reliability and Maintenance Optimization
- Fatigue and fracture mechanics
- Elevator Systems and Control
- Machine Learning and ELM
Chengdu University of Traditional Chinese Medicine
2023-2024
Wannan Medical College
2024
Beihang University
2014-2023
Xi’an Jiaotong-Liverpool University
2023
Dalian University of Technology
2023
University of Illinois Urbana-Champaign
2023
Xi'an Jiaotong University
2016-2021
China XD Group (China)
2021
Nanjing University of Chinese Medicine
2020
Southwest Jiaotong University
2019
The key challenge of intelligent fault diagnosis is to develop features that can distinguish different categories. Because the unique properties mechanical data, predetermined based on prior knowledge are usually used as inputs for classification. However, proper selection often requires expertise and becomes more difficult time consuming when volume data increases. In this paper, a novel deep learning network (LiftingNet) proposed learn adaptively from raw without knowledge. Inspired by...
Data-driven intelligent diagnosis models expect to mine the health information of machines from massive monitoring data. However, size faulty data collected in engineering scenarios is limited, which leads few-shot fault as a valuable research point. Fortunately, it possible reduce required amount training by integrating prior knowledge into models. Inspired this, we present knowledge-augmented self-supervised feature learning framework for diagnosis. In framework, 24 signal indicators are...
The key to intelligent fault diagnosis is find relevant characteristics with the capability of representing different types faults. However, engineering problem that a few simple empirical features (EFs) cannot obtain high classification accuracy, and complex feature requires strong professional knowledge, which leads limited applications on general scale. In addition, extraction methods without prior knowledge guarantee model learned used for classification, its robustness generalization...
Intelligent fault detection is an important application of artificial intelligence and has been widely used in many mechanical systems. The shipborne antenna that a typical system plays irreplaceable role ships. Considering the tough working environment heavy background noise, difficult for antenna. Therefore, paper presents intelligent method via multiscale inner product with locally connected feature extraction detection. Inspired by principle, this takes advantage to capture information...
Class imbalance issue has been a major problem in mechanical fault detection, which exists when the number of instances presenting class is significantly fewer than that another class. This article focuses on zero-shot detection rolling bearing, extreme case imbalance. Aiming at this problem, two-stage recognition method proposed. First, inspired by conditional generative adversarial network, novel feature generating network composed extractor, discriminator, and generator designed to...
This study focuses on the machined surface integrity of titanium alloy Ti-10V-2Fe-3Al (Ti-1023) during face milling. Surface roughness, machining defect, microhardness, and microstructure variations are investigated at different cutting speeds tool average flank wear (VB) values. Experimental results show that roughness increases when speed is increased from 40 to 100 m min−1 decreases 300 by using a new tool. Moreover, values stable worn VB = 0.2 mm increased. As for defects, defect...
Magnetic sensors based on tunneling magnetoresistance (TMR) effect exhibit high sensitivity, small size, and low power consumption. They have gained a lot of attention potential applications in various domains. This study first introduces the development history basic principles TMR sensors. Then, comprehensive description linearization Wheatstone bridge configuration is presented. Two key performance parameters, field sensitivity noise mechanisms, are considered. Finally, emerging discussed.
Fault diagnosis is vital to ensuring the security of rotating machinery operations. While fault data obtained from mechanical equipment for this issue are often insufficient and no labels. In case, supervised algorithms cannot come into play. Hence, article proposes a self-supervised simple Siamese framework (SSF) bearing based on contrastive learning algorithm SimSiam which uses simplified network find distinguishable features different categories. SSF consists weight-sharing encoder...
Intelligent fault detection has been widely used for feature extraction and classification. However, various complex signal processing methods are adopted in many researches. This article presents a novel deep learning network via shunt-wound restricted Boltzmann machines (RBMs) with layerwise to learn the features from big raw vibration signals directly. The consists of split layer, predict update dephasing softmax layer. RBMs both layer improve ability ensure effective training network....
Deep learning has demonstrated splendid performance in mechanical fault diagnosis on condition that source and target data are identically distributed. In engineering practice, however, the domain shift between domains significantly limits further application of intelligent algorithms. Despite various transfer techniques proposed, either they focus single-source adaptation (SDA) or utilize multisource globally locally, which both cannot address cross-domain effectively, especially with...