A5034050336- Machine Fault Diagnosis Techniques
- Aluminum Alloy Microstructure Properties
- Fault Detection and Control Systems
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Metallurgy and Material Forming
- Aluminum Alloys Composites Properties
- Anomaly Detection Techniques and Applications
- Advanced Adaptive Filtering Techniques
- Acoustic Wave Phenomena Research
- Advanced Materials Characterization Techniques
- Mechanical Engineering and Vibrations Research
- Tribology and Lubrication Engineering
- Risk and Safety Analysis
- Advanced Combustion Engine Technologies
- Mineral Processing and Grinding
- Advancements in Battery Materials
- Speech and Audio Processing
- Construction Project Management and Performance
- Additive Manufacturing Materials and Processes
- Education and Work Dynamics
- Grouting, Rheology, and Soil Mechanics
- Civil and Geotechnical Engineering Research
- Digital Filter Design and Implementation
- Oil and Gas Production Techniques
Hebei University of Technology
2022-2024
Hebei University of Science and Technology
2023
London South Bank University
2018-2022
Shanghai Normal University
2011
To improve the ability of deep learning model to handle imbalanced data, a fault diagnosis method based on improved gated convolutional neural network (IGCNN) is proposed. Firstly, an convolution layer proposed for feature extraction, with batch normalisation (BN) applied adjust data distribution and enhance generalisation performance model. Then, learned by multiple layers pooling fed fully connected type identification. Finally, label-distribution-aware margin (LDAM) loss function employed...
Machine learning, especially deep has been highly successful in data-intensive applications, however, the performance of these models will drop significantly when amount training data does not meet requirement. This leads to so-called Few-Shot Learning (FSL) problem, which requires model rapidly generalize new tasks that containing only a few labeled samples. In this paper, we proposed model, called convolutional meta-learning networks (DCMLN), address low generalization under limited for...
Pumps are one of the most critical machines in petrochemical process. Condition monitoring such parts and detecting faults at an early stage crucial for reducing downtime production line improving plant safety, efficiency reliability. This paper develops a fault detection isolation scheme based on unsupervised machine learning method, sparse autoencoder (SAE), evaluates model industrial multivariate data. The Mahalanobis distance (MD) is employed to calculate statistical difference residual...
The marine engine is a complex-structured multidisciplinary system that operates in harsh environment involving high temperatures and pressures gas/fluid/solid interactions. Many malfunctions faults can occur to the efficient condition monitoring critical ensure expected performance. In this paper, test rig established its process data are recorded, including various pressures. Two data-driven models, i.e., principal component analysis sparse autoencoder, physics-based model applied for two...
<title>Abstract</title> The vibration signals of a planetary gear set exhibit rich spectra that can reflect its operating states. In previous phenomenological models, the sidebands are mostly explored under condition constant speed. However, signal is not only affected by internal nonlinear factors such as velocity fluctuations, fault excitation and manufacturing errors, but also transmission path effects cause to be further modulated, making it difficult recognize characteristics. To solve...
Substation noise is a crucial factor that influences residents’ quality of life, especially in the densely residential areas. Despite small- and medium-sized transformer facilities having relatively low levels, due to their proximity areas, they generate considerable annoyance, rendering them focal point among environmental complaints. The predominant emitted by these falls within medium- low-frequency spectrum range, conventional passive reduction techniques exhibit limited efficacy...
Large rotating machinery, such as centrifugal gas compressors and pumps, have been widely applied acted crucial components in the oil industries. Breakdowns or deteriorated performance of these machines can bring significant economic loss to companies. In order conduct effective maintenance avoid unplanned downtime, a system-wide health indicator is proposed this paper. The not only uses dynamic risk profile, but also considers financial fault probability based on condition monitoring data....
Planetary gearbox is numerously used in various mechanical transmission systems because of their extensive bearing range and high reliability. However, due to limitations manufacturing technique economic considerations, the gear tooth surface roughness inevitable exists practical process. To analyze influence on vibration signals a planetary system, nonlinear dynamic model considering multi-factor coupling established. The takes into account roughness, backlash, time-varying meshing...