A5100705104- Machine Fault Diagnosis Techniques
- Gear and Bearing Dynamics Analysis
- Fault Detection and Control Systems
- Advanced Measurement and Detection Methods
- Structural Health Monitoring Techniques
- Industrial Vision Systems and Defect Detection
- Infrastructure Maintenance and Monitoring
- Advanced Sensor and Control Systems
- Mineral Processing and Grinding
- Image Processing Techniques and Applications
- Control Systems in Engineering
- Speech and Audio Processing
- Advanced machining processes and optimization
- Acoustic Wave Phenomena Research
- Hydraulic and Pneumatic Systems
- Metallurgical Processes and Thermodynamics
- Advanced Decision-Making Techniques
- Advanced Algorithms and Applications
- Target Tracking and Data Fusion in Sensor Networks
- Manufacturing Process and Optimization
- Image and Object Detection Techniques
- Blind Source Separation Techniques
- Mechanical Failure Analysis and Simulation
- Neural dynamics and brain function
- Image and Signal Denoising Methods
Kunming University of Science and Technology
2015-2025
Kunming University
2025
National Taipei University of Technology
2023-2024
Wuhan University of Science and Technology
2017-2024
China University of Mining and Technology
2024
Lanzhou Jiaotong University
2021-2022
Air Force Institute of Aviation Medicine Affiliated Hospital
2012-2022
Shaanxi University of Science and Technology
2017-2022
Guilin University of Electronic Technology
2019-2021
Liaoning Shihua University
2009-2020
Aiming at the problem that a single neural network model has difficulty in accurately predicting trends of remaining useful life rolling bearing, method bearings using gated recurrent unit-deep autoregressive (GRU-DeepAR) with an adaptive failure threshold was proposed. First, time domain and frequency features were extracted from bearing vibration signal. Second, its operation process divided into smooth stage degradation according to trend accumulated root mean square maximum. Then, for...
The characteristic of vibration signal bearing compound fault CNC machine tool is that the are not clear, and it hard to separate extract error. An excellent approach for extracting periodic pulses from signals multi-point optimal minimum entropy deconvolution adjustment (MOMEDA). Nevertheless, chosen pulse cycle filter length strongly influence performance MOMEDA. To address some these shortcomings MOMEDA, a method optimizing MOMEDA parameters based on kurtosis composite index proposed....
Traditional bearing fault diagnosis mainly based on periodic feature extraction from pulses. However, the diagnostic accuracy was limited due to absence of rotation speed information. Firstly, through envelope analysis and order tracking extract roller passing frequency (RPF) order, this paper verifies performance vibration acoustic emission signals (AE) in low-speed. Secondly, advantages source localization AE are also verified analyzed. In accurately pulse generated by source, a method for...
In light of the problems a single vibration feature containing limited information on degradation rolling bearings, redundant in high-dimensional sets inaccurately reflecting reliability bearings service, and assessments performance being disturbed by outliers false fluctuations signal, this study proposes method assessing bearings’ terms using adaptive sensitive selection multi-strategy optimized support vector data description (SVDD). First, set signals from was extracted. Second,...
Abstract As a core component of industrial robots, the RV reducer directly affects normal operation robot, so it is great significance to monitor its status and diagnose faults. In field fault diagnosis, intelligent diagnosis methods based on deep learning have shown advantages in accuracy efficiency. However, as network depth scale increase, exponentially growing model computation parameter amounts require higher hardware requirements for computers, making difficult deploy embedded...
Displacement measurement is an essential method for structural safety assessment and health monitoring, the static dynamic characteristics of structure can be obtained through displacement. In order to overcome limitations sensors in vibration large structures, as well poor adaptability visual algorithms such machine learning digital image processing, this paper takes bridge research object introduces deep into field measurement. Moreover, based on convolutional neural network, a new...
Abstract Rolling element bearings are critical components in rotating machinery. To tackle the problem of difficult to accurately characterize operating state rolling caused by irrelevance and varying sensitivity multiple features performance degradation, introduction subjective errors determination hyperparameters deep learning models, which can affect accuracy efficiency remaining useful life (RUL) prediction. address these challenges, this paper proposed a novel RUL prediction method for...
Condition monitoring and fault diagnosis are the most important process in manufacturing industries. In this paper, a steel beam panoramic crack detection method based on structured random forests has been proposed to obtain more efficient maintenance of equipment. The semi-reconstruction anti-symmetrical bi-orthogonal wavelets combined detect edges cracks. Candidate features images randomly chosen train classifier. Besides, fast-multi-image stitching is applied stitch entire image....
The motor-gearbox systems are widely used in industrial production, and the variations of gear meshing stiffness generate transient frequency modulation (FM) effects on motor current. second-order synchrosqueezing transform (SST) is effective for revealing time-varying characteristics current, thus enabling monitoring condition electromechanical system. However, selection window parameter SST analysis subjective lacks specific theoretical guidance. To quantify analytical limits, this study...