A5069886339- Machine Fault Diagnosis Techniques
- Gear and Bearing Dynamics Analysis
- Fault Detection and Control Systems
- Engineering Diagnostics and Reliability
- Industrial Vision Systems and Defect Detection
- Non-Destructive Testing Techniques
- Anomaly Detection Techniques and Applications
- Advanced Measurement and Detection Methods
- Structural Health Monitoring Techniques
- Advanced machining processes and optimization
- Structural Integrity and Reliability Analysis
- Tribology and Lubrication Engineering
- Ultrasonics and Acoustic Wave Propagation
- Blind Source Separation Techniques
- Domain Adaptation and Few-Shot Learning
- Advanced Decision-Making Techniques
- Industrial Technology and Control Systems
- Mechanical Failure Analysis and Simulation
- Machine Learning and ELM
- Image Processing Techniques and Applications
- Mechanical stress and fatigue analysis
- Evaluation and Optimization Models
- Electronic and Structural Properties of Oxides
- Advancements in Solid Oxide Fuel Cells
- Peptidase Inhibition and Analysis
Shandong University of Science and Technology
2020-2025
Nanjing University of Aeronautics and Astronautics
2019-2023
Zhengzhou University
2018
Domain adaptation (DA)-based methods for fault diagnosis (FD) of rotating machinery have achieved impressive results in recent years. Most hold the assumption that source domain (SD) and target (TD) share same label space, which is not always satisfied actual situations. A more practical scenario called partial (PDA) needs to be given attention, where transferable knowledge learned from a larger SD applied smaller but relevant TD. PDA method class-weighted alignment-based transfer network...
Abstract Transfer learning has been successfully applied in fault diagnosis to solve the difficulty constructing network models due lack of labeled data practical engineering. The current transfer mainly use adaptive method obtain similarity between source and target domains, but obtained is incomplete. Inspired by domain-adversarial mechanism, a novel called ‘distance guided network’ (DGDAN) proposed this study. DGDAN includes two modules: maximum mean discrepancies (MMD) domain adaptation....
Abstract Random impact interference has always been an important subject in fault diagnosis. In the fast kurtogram (FK), kurtosis is a sparsity index used to locate optimal resonance frequency bands. However, shortcomings of measure are tendency select random with large amplitude and relatively weak sparse ability, which reduces accuracy anti-interference FK method. To overcome effects interference, this paper proposes novel bearing diagnosis method named nonlinear Hoyergram (FNH), Hoyer...
Existing generative adversarial networks (GAN) have potential in data augmentation and the intelligent fault diagnosis of bearings. However, most relevant studies only focus on rotating machines with sufficient fault-type samples, some rare samples may be missing training practical engineering. To address those deficiencies, this paper presents an method based dynamic simulation model Wasserstein network gradient normalization (WGAN-GN). The bearing faults is constructed to obtaining signals...
Abstract Distance-based domain adaptation methods have received extensive application in the transfer learning field. Different distances different characteristics due to various data processing principles. Therefore, choosing appropriate distance can accomplish tasks more efficiently. Domain adversarial neural networks extract invariant features through game confrontation, but it is not capable of extracting hidden gear under speed fluctuations, and only using mechanism for feature...
Fault diagnosis of rotating machinery has always drawn wide attention. In this paper, Intrinsic Component Filtering (ICF), which achieves population sparsity and lifetime consistency using two constraints: l1/2 norm column features l3/2 -norm row features, is proposed for the fault diagnosis. ICF can be used as a feature learning algorithm, learned fed into classification to achieve automatic classification. also filter training method extract separate weak components from noise signals...
Abstract The generalized nonlinear sparse spectrum (GNSS), as an improved fast kurtogram (FK) method, effectively suppresses the interference of abnormal signals through preprocessing and enhancement. However, GNSS method inherits shortcoming traditional FK using finite impulse response filters to process nonstationary signals, which limits accuracy fault extraction. Therefore, more precise should be developed further improve performance features. Inspired by this, this paper introduces...
Domain adaptive fault diagnosis methods of bearing have made extraordinary achievements in recent years. Among these methods, the vast majority machine learning problems are balanced classification problems. However, imbalance very common practical applications. Undersampling and oversampling often used to deal with Inspired by Wasserstein generative adversarial network (WGAN), a domain guided WGAN for under imbalanced samples is proposed this article. First, unbalanced data each collected,...
Abstract Bearing faults under different operating conditions often cannot be diagnosed by models trained a single operational condition. Additionally, the extraction of domain-invariant features in domain adaptation (DA) algorithms is also challenge. To address aforementioned issues, multi-layer model based on an improved sparse autoencoders (SAEs) and dual-domain distance mechanism (ISAE-DDM) proposed. First, feature capability traditional SAEs enhanced strategy that combines mean squared...
Abstract This paper presents a comprehensive review of recent advancements in bearing health monitoring and remaining useful life (RUL) prediction. It highlights key innovations anomaly detection, indicator (HI) construction, degradation modeling, RUL estimation, examining developments across statistical, machine learning, deep learning approaches while analyzing their strengths, limitations, application contexts. Special emphasis is placed on the role capturing complex patterns from...
Abstract Fault diagnosis is of significance for ensuring the safe and reliable operation machinery equipment. Due to heavy noise interference, it difficult detect fault directly from measured signal. Hence, signal processing techniques that can achieve feature extraction denoising are most common tools in field. The Adaptive maximum second-order cyclostationarity blind deconvolution (ACYCBD), an advanced technique, algorithm shows great superiority reduction faulty signals. However, ACYCBD...