A5100456613- Machine Fault Diagnosis Techniques
- Gear and Bearing Dynamics Analysis
- Structural Health Monitoring Techniques
- Fault Detection and Control Systems
- Engineering Diagnostics and Reliability
- Advanced Algorithms and Applications
- Advanced machining processes and optimization
- Advanced Decision-Making Techniques
- Vacuum and Plasma Arcs
- Power Transformer Diagnostics and Insulation
- Electronic and Structural Properties of Oxides
- Electrical Contact Performance and Analysis
- Power System Reliability and Maintenance
- Ferroelectric and Piezoelectric Materials
- Evaluation and Optimization Models
- Occupational Health and Safety Research
- Smart Grid and Power Systems
- Non-Destructive Testing Techniques
- Image and Signal Denoising Methods
- Dielectric properties of ceramics
- Mechanical Failure Analysis and Simulation
- Welding Techniques and Residual Stresses
Chang'an University
2024
Beijing Information Science & Technology University
2024
Tongji University
2023
Shenyang Institute of Engineering
2023
Beijing University of Technology
2021-2022
Beijing Aerospace Flight Control Center
2022
State Grid Nanjing Power Supply Company (China)
2019
North China Electric Power University
2019
Ministry of Education of the People's Republic of China
2003
Abstract When rolling bearings work under variable speed conditions and experience compound faults, their vibration signal components become more complex, some of which coexist in a narrow frequency band, are easy to interfere with each other coupled together. In cases, the tachometer cannot or is difficult install, resulting insufficient information, further increasing challenge fault diagnosis. Aiming at above problem diagnosis without , method based on adaptive chirp decomposition...
The vibration signals produced by rotating machinery are mostly non-stationary, and there numerous methods for dealing with them. People's expectations time–frequency analysis (TFA) results increasing all the time. emergence of post-processing algorithms based on short-time Fourier transform (STFT) provides scholars new ideas, but such heavily rely window length selected STFT have significant uncertainty. To address this issue, we propose generalized S-synchroextracting transform, a...
In the condition evaluation of high-voltage SF6 circuit breaker, contact resistance and mass loss have a significant impact on arc contact. To that end, this paper proposes method based quantum particle swarm optimization support vector regression (QPSO-SVR), implementation which can effectively predict increment breaker contacts under different current conditions, best (SVR) algorithm training parameters are obtained through experimental data. validate proposed method's accuracy, it is...
Abstract Bearings of wind turbines have become one the components with high failure rate in because their bad operating environment. In this paper, a fault diagnosis model based on deep belief network is proposed for bearing turbine. The time-frequency spectrum turbine vibration data after short-time Fourier transform (STFT) used as input model, and output identification code various types bearing. Compared time domain signal to bearings, has higher recognition accuracy. Based different...
Deep learning (DL) has inaugurated new approaches to implement fault diagnosis of roller bearings, which is essential deal with the current industrial big data era. Unfortunately, many existing deep models, particularly convolutional neural network (CNN) have following drawbacks. Single-channel convolution in pooling layer problem loss. The feature information extracted by CNN not integrated due lack ability, will induce unsatisfactory diagnostic results and poor generalization ability. To...
Since the actual operation of bearing inevitably exists in both noise and variable working conditions, most traditional networks can only deal with them alone, fault identification result will be significantly reduced under such complex conditions. Therefore, Parallel Multichannel Deep Convolution Neural Network (PMDCNN) Long Short-Term Memory (LSTM) are proposed as PMDCNN-LSTM model to enable better performance. And a local sparse structure is used greatly reduce number parameters,...
It is difficult to realize the effective separation of signal and noise in vibration signals, owing traditional filtering method. This paper puts forward an algorithm which was based on Hasudorff distance nonlocal means (Hasudorff-Non Local Means, HDNLM). makes use more efficient than Euclidean preserving details. aim improve weight model NLM Distance. Experimental simulation show that: The improved superior simple morphology, NLM, db8 wave transform denoising aspect. newly method combined...
Accurate fault diagnosis is critical because it has a great impact on operational stability of hypersonic vehicles. Recent trends various literatures shows that deep learning promising methodology to tackle many challenging tasks. In this study, intelligent method based network proposed for diagnosis. The constructed by serial coupling the one-dimensional Residual Convolution neural networks with Attention mechanism (ResCNN-ATT) and Long short-term memory (LSTM-ATT), which referred as LSTM...
Abstract AC contactors are widely used in various low voltage electrical control circuits. The operation status of contactor will affect the safety and reliability power system operation, so it is important to accurately evaluate contactor. An outstanding feature intelligence that can be monitored on-line, which great significance for evaluating performance degradation contactors. Accurate acquisition physical quantities such as speed acceleration difficult at present. Traditional method...
The time-frequency representation (TFR) is the result of processing signal by analysis method, which can reveal time-varying characteristics signal. TFR obtained post-processing algorithm has energy accumulation and high resolution. generalized S-synchroextracting transform (GS-SET) stands out for its strong adaptability. However, when this method deals with signals whose instantaneous frequencies are close to each other or cross other, will be indistinguishable vague. To solve problem, a...