A5015560656- Machine Fault Diagnosis Techniques
- Machine Learning and ELM
- Domain Adaptation and Few-Shot Learning
- Anomaly Detection Techniques and Applications
- Structural Health Monitoring Techniques
- Advanced Electrical Measurement Techniques
- Reliability and Maintenance Optimization
- Gear and Bearing Dynamics Analysis
- Fault Detection and Control Systems
Shanghai Jiao Tong University
2022-2023
Existing cross-network node classification methods are mainly proposed for closed-set setting, where the source network and target share exactly same label space. Such a setting is restricted in real-world applications, since might contain additional classes that not present source. In this work, we study more realistic open-set (O-CNNC) problem, contains all known further several target-private unseen Borrowing concept from domain adaptation, defined as an “unknown” class. To address...
Existing cross-network node classification methods are mainly proposed for closed-set setting, where the source network and target share exactly same label space. Such a setting is restricted in real-world applications, since might contain additional classes that not present source. In this work, we study more realistic open-set (O-CNNC) problem, contains all known further several target-private unseen Borrowing concept from domain adaptation, defined as an unknown class. To address...
Time–frequency analysis (TFA) methods can characterize the instantaneous frequency (IF) features of nonstationary signals by constructing time-domain signal model. However, traditional TFA are ineffective for transient that discontinuous in time domain and characterized sharp variation. Therefore, this article proposes a new method called generalized transient-squeezing transform (GTST) to process model domain. First, group delay (GD) estimation is calculated based on frequency-domain...
Rolling bearings are widely applied in rotating machinery, and its performance degradation assessment is a crucial work. Accurate remaining useful life (RUL) prediction of great significance for devising maintenance scheme preventing sudden shutdown with the selection representative features being key factor. In this paper, novel model constructed based on signal index (SSI) deep cumulative (DCF). First, succinct fast empirical mode decomposition (SFEMD) employed to decompose original...