A5100784774- Machine Fault Diagnosis Techniques
- Fault Detection and Control Systems
- Engineering Diagnostics and Reliability
- Smart Agriculture and AI
- Gear and Bearing Dynamics Analysis
- Hand Gesture Recognition Systems
- Spectroscopy and Chemometric Analyses
- Muscle activation and electromyography studies
- Laser-induced spectroscopy and plasma
- Analytical chemistry methods development
- Gait Recognition and Analysis
- Anomaly Detection Techniques and Applications
- Cultural Heritage Materials Analysis
- Remote Sensing and Land Use
- Ultrasonics and Acoustic Wave Propagation
- Structural Integrity and Reliability Analysis
- Advanced Measurement and Detection Methods
- Wireless Sensor Networks and IoT
- Gaze Tracking and Assistive Technology
- EEG and Brain-Computer Interfaces
- Mercury impact and mitigation studies
- Shoulder Injury and Treatment
Hunan University
2014-2024
Minzu University of China
2023-2024
Xi'an Jiaotong University
2020-2022
Shanghai University
2019-2021
Henan University of Science and Technology
2018-2019
China Agricultural University
2012
Digital image recognition of plant diseases could reduce the dependence agricultural production on professional and technical personnel in protection field is conducive to development informatization. In order find out a method realize diseases, four kinds neural networks including backpropagation (BP) networks, radial basis function (RBF) generalized regression (GRNNs) probabilistic (PNNs) were used distinguish wheat stripe rust from leaf grape downy mildew powdery based color features,...
The planetary gearbox, widely used in many machinery fields, suffers from harmful vibration excited by bearings fault, which always causes machine breakdowns. Thus, fault diagnosis is the necessary approach to keeping machines safe, often takes features, are extracted signals, as critical information. However, limited various interferences caused gear meshing and background noise, characteristic information weak difficult identify, bringing an increasing emphasis on feature enhancement. In...
Traction motor bearings are crucial components to guarantee stable operation. Thus, it is significant monitor the bearing condition. Dictionary learning a powerful method extract characteristic Compared with ordinary dictionary learning, multiscale applied transform coefficients performs well in extracting impact signals and takes less time. However, spends more time tuning parameters make algorithm efficient, especially practical industrial applications. Hence, faster adaptive parameter...
Most of the traditional fault diagnosis methods rely on expert knowledge artificial extraction features and related fields, these algorithms are not accurate, robustness generalization ability poor. Convolutional neural network is one most widely used deep learning models. Based its unique convolution-pooling structure, convolutional has powerful feature expression capabilities. In this paper, based characteristics one-dimensional vibration signals, a algorithm model proposed. Through...
Maize leaf disease is a common problem in the process of crop growth. It great significance to identify and monitor timely accurately for benefit yield agricultural production. This paper aims propose method maize recognition based on improved Faster R-CNN. First, Region Proposal Network (RPN) introduced R-CNN generate candidate regions reduce computational load identification process. Secondly, network uses Adam optimizer optimize training model improve convergence speed performance....
Most of the traditional fault diagnosis methods rely on artificial extraction features and expert knowledge related fields, these algorithms are not accurate, ability robustness generalization poor. Convolutional neural network is one most widely used deep learning models. Based its unique convolution-pooling structure, convolutional has powerful feature expression capabilities. In this paper, structure classical optimized, a algorithm based optimization for proposed. For characteristics...