Special Issues
Table of Content

Data-Driven and Physics-Informed Machine Learning for Digital Twin, Surrogate Modeling, and Model Discovery, with An Emphasis on Industrial Applications

Submission Deadline: 01 April 2025 View: 145 Submit to Special Issue

Guest Editors

Prof. Ki-Yong Oh

Email: kiyongoh@hanyang.ac.kr

Affiliation: School of Mechanical Engineering, Hanyang University

Homepage:

Research Interests: AI Transformation, Applied Dynamics, Automation and Optimization in Smart factory Physics-informed Neural Network, Prognosis and Health Management

图片3.png


Prof. Hyunseok Oh

Email: hsoh@gist.ac.kr

Affiliation: School of Mechanical Engineering, Gwangju Institute of Science and Technology

Homepage: sddo.gist.ac.kr

Research Interests: Physics-informed machine learning, Industrial artificial intelligence, Fault diagnostics and prognostics, Design optimization

图片4.png


Prof. Minseok Choi

Email: mchoi@postech.ac.kr

Affiliation: Mathematics, POSTECH

Homepage:

Research Interests: scientific machine learning, AI for science, uncertainty quantification 

图片5.png


Prof. Jae Hyuk Lim

Email: jaehyuklim@jbnu.ac.kr

Affiliation: Department of Mechanical Engineering, Jeonbuk National University

Homepage:

Research Interests: Data-driven and physics-informed machine learning for digital twin, surrogate modeling, and model discovery, with an emphasis on industrial applications.

图片6.png


Summary

1) Introduction

The intersection of data-driven techniques and physics-informed machine learning is revolutionizing the landscape of computational modeling and simulation. These advancements are particularly significant in the context of digital twins, surrogate modeling, and model discovery, offering new pathways to optimize and understand complex industrial systems. The integration of data and physics-informed methods leverages the strengths of both approaches, resulting in robust, accurate, and computationally efficient models that are highly relevant for various industrial applications.


2) Objectives of the Special Issue

This special issue aims to showcase cutting-edge research and practical implementations of data-driven and physics-informed machine learning methods in the realms of digital twins, surrogate modeling, and model discovery. The focus will be on how these innovative approaches are being applied to solve real-world industrial problems, enhancing efficiency, reducing costs, and improving predictive capabilities.


3) The special issue will cover a wide range of topics related to the development and application of PINNs in multiphysics systems, including but not limited to:


· Digital Twins:

- Development and implementation of digital twins using machine learning and physics-informed approaches.

- Real-time monitoring, prediction, and optimization using digital twins.

- Case studies of digital twin applications in various industries (e.g., manufacturing, aerospace, automotive).


· Surrogate Modeling:

- Creation of surrogate models for complex systems using data-driven techniques.

- Integration of physics-based knowledge into surrogate models to enhance accuracy and generalization.

- Applications of surrogate models in design optimization, uncertainty quantification, and real-time simulation.


·  Model Discovery:

- Techniques for discovering governing equations and underlying physics from data.

- Data-driven approaches to uncover hidden patterns and insights in industrial processes.

- Combining machine learning with domain knowledge for model discovery and validation.


·  Industrial Applications:

- Applications in manufacturing, energy systems, aerospace, automotive, and other industrial sectors.

- Case studies demonstrating the impact of data-driven and physics-informed models on industrial operations.

Challenges and opportunities in deploying these models in real-world industrial settings.


Keywords

Data-driven Machine Learning, Physics-Informed Machine Learning, Digital Twin, Surrogate Modeling, Model Discovery, Industrial Applications, Real-time Monitoring, Predictive Maintenance, Design Optimization, Uncertainty Quantification, Real-time Simulation, Domain Knowledge Integration, Industrial Process Optimization

Share Link