Innovative Research Award
DongHee Park
DAVISS, South Korea
DongHee Park is a researcher affiliated with DAVISS in South Korea whose documented scholarly work applies data science and machine-learning methods to industrial fault diagnosis. The available record contains eight publications addressing condition-index representations, diagnostic automation, convolutional neural networks and feature classification for machinery faults. [1]
Abstract
Park’s research focuses on computational approaches for identifying and classifying faults in industrial machinery. Recent work compares conventional feature-based representations with physics-guided condition-index representations for motor fault diagnosis, while related studies investigate automated elevator diagnosis, rotor-system classification and machine-learning methods for blade-rubbing defects. [1] [2] The research demonstrates a consistent interest in translating operational data into diagnostic information.
Keywords
Data science; machine learning; fault diagnosis; condition monitoring; industrial motors; diagnostic automation; convolutional neural networks; rotor systems; blade rubbing; feature classification; predictive maintenance.
Introduction
Modern industrial systems generate large quantities of operational data that can be analysed to detect abnormal behaviour. Machine-learning algorithms and engineered diagnostic features provide methods for distinguishing normal operation from specific fault conditions. Park’s publication record addresses these challenges across several machinery categories, linking data analytics with practical condition-monitoring problems. [3]
Research Profile
The supplied profile records eight documents in the subject area of Data Science and Analytics. Citation and h-index values were not provided and therefore are not inferred. The publication sequence from 2024 to 2026 indicates sustained investigation of machine-learning feature development, classification methods, automated diagnosis and physics-guided representations.
Research Contributions
- Evaluation of conventional and physics-guided representations for industrial motor fault diagnosis. [1]
- Development of rule-based automation technology for elevator fault diagnosis. [2]
- Use of combination images of feature vectors with convolutional neural networks for rotor fault classification. [3]
- Development and classification of machine-learning features for early blade-rubbing diagnosis. [4] [5]
Publications
Comparative Evaluation of Conventional Feature-Based and Physics-Guided Condition-Index Representations for Industrial Motor Fault Diagnosis (2026), Machines. [1]
Development of Rule-Based Diagnostic Automation Technology for Elevator Fault Diagnosis (2025), Sensors. [2]
CNN-based fault classification using combination image of feature vectors in rotor systems (2024), Journal of Mechanical Science and Technology. [3]
A Study on Machine Learning-Based Feature Classification for the Early Diagnosis of Blade Rubbing (2024), Sensors.[4]
Development of features for blade rubbing defect classification in machine learning (2024), Journal of Mechanical Science and Technology.[5]
Research Impact
Because citation and h-index values were not supplied, quantitative scholarly impact cannot be assessed from the available profile alone. The publication record nevertheless demonstrates application of data-driven techniques to multiple industrial systems, including motors, elevators, rotor systems and machinery affected by blade rubbing. Publication across Machines, Sensors and the Journal of Mechanical Science and Technology provides a documented scholarly context for these contributions. [1] [3]
Award Suitability
The Innovative Research Award profile is supported by a coherent body of research applying machine learning and analytical representations to industrial fault diagnosis. The progression from feature development and classification toward automated diagnostic systems and physics-guided condition indices indicates methodological continuity. [2] [5] Final recognition should be determined using the complete scholarly record, originality, methodological quality, practical relevance, reproducibility and broader research contributions.
Conclusion
DongHee Park’s documented research is centred on data science and analytics for industrial condition monitoring and fault diagnosis. Publications from 2024 through 2026 cover machine-learning features, convolutional neural networks, rule-based automation and physics-guided diagnostic representations. This focused research trajectory provides a substantive basis for consideration for the Innovative Research Award within the International Soil Scientist Awards framework.
External Links
References
- Park, DongHee. (2026). Comparative Evaluation of Conventional Feature-Based and Physics-Guided Condition-Index Representations for Industrial Motor Fault Diagnosis. Machines.
https://doi.org/10.3390/machines14091024
- Park, DongHee. (2025). Development of Rule-Based Diagnostic Automation Technology for Elevator Fault Diagnosis. Sensors.
https://doi.org/10.3390/s26010223
- Park, DongHee. (2024). CNN-based fault classification using combination image of feature vectors in rotor systems. Journal of Mechanical Science and Technology.
https://doi.org/10.1007/s12206-024-1006-z
- Park, DongHee. (2024). A Study on Machine Learning-Based Feature Classification for the Early Diagnosis of Blade Rubbing. Sensors.
https://doi.org/10.3390/s24186013
- Park, DongHee. (2024). Development of features for blade rubbing defect classification in machine learning. Journal of Mechanical Science and Technology.
https://doi.org/10.1007/s12206-023-1201-3