Xueyun Liu
Papers
1
Total Citations
84
H-Index
1
About
Xueyun Liu is a leading researcher in intelligent fault diagnosis and machine learning for industrial systems, with a focus on addressing data imbalance challenges in critical infrastructure. Her most influential work, "Feature-level SMOTE: Augmenting fault samples in learnable feature space for imbalanced fault diagnosis of gas turbines" (2023), has garnered 84 citations, establishing her as a key innovator in synthetic data augmentation techniques. Liu’s major contribution lies in developing a feature-level adaptation of the Synthetic Minority Oversampling Technique (SMOTE), which generates realistic fault samples within a learned feature space rather than raw data, significantly improving diagnostic accuracy for rare but critical failure modes in gas turbines. This work has direct implications for predictive maintenance, reducing downtime and enhancing safety in energy and aerospace sectors. Her research bridges the gap between theoretical machine learning and practical industrial applications, offering scalable solutions for imbalanced datasets common in real-world monitoring. Liu’s achievements highlight her ability to translate complex methodological advances into tangible engineering outcomes, making her a vital figure in the evolution of intelligent condition monitoring systems.
Research Focus
Key Achievements
Top Papers
- 1