Submission Deadline: 31 March 2027 View: 133 Submit to Special Issue
Prof. Dr. Hamid Reza Karimi
Email: hamidreza.karimi@polimi.it
Affiliation: Department of Mechanical Engineering, Politecnico di Milano, Milano, Italy
Research Interests: fault diagnosis, fault prognosis, machine learning, vibration control, condition monitoring
Prof. Dr. Funa Zhou
Email: zhoufn@shmtu.edu.cn
Affiliation: School of Logistic Engineering, Shanghai Maritime University, Shanghai, China
Research Interests: fault diagnosis, fault prognosis, machine learning, vibration control, condition monitoring
Modern control and monitoring systems rely heavily on modeling techniques to capture unknown system dynamics, making accurate modeling a cornerstone of control theory and engineering practice. Traditional approaches are generally classified as physics-driven or data-driven. Physics-driven methods offer interpretability and a foundation in system laws but often struggle with uncertainties and complex behaviors. Data-driven methods can adapt to complexity but are limited by data quality, quantity, and variability, reducing robustness and generalization.
Data–physical driven methods have emerged as a promising approach by integrating physical knowledge with data-driven learning. This synergy enhances interpretability, robustness, and adaptability, addressing the limitations of purely physics- or data-driven models. Such methods are particularly valuable for health monitoring, fault diagnosis, and fault prognosis, where accurate modeling, uncertainty handling, and limited fault data present critical challenges.
Despite advances, key issues remain, including adherence to physical constraints, accurate sensing and signal fusion, and the need for interdisciplinary expertise. This Special Issue provides a platform for researchers to share cutting-edge developments in data–physical driven modeling, control, and health monitoring. Topics include, but not limited to, digital twin-based modeling, filtering, deep learning, reinforcement learning, fault detection and prognosis, and industrial applications, aiming to advance both theoretical foundations and practical implementations.
Login
Register
Submit a Paper
Propose a Special lssue