BWTC- Bridging Physics and Machine Learning: Real-World Applications
Abstract- The integration of physics-informed machine learning (PIML) represents a paradigm shift in scientific and engineering research, offering a powerful framework to address complex, real-world problems by embedding physical laws into data-driven models. Investigating the synergy between traditional physics-based approaches and advanced machine learning techniques is important to enhance model efficiency, interpretability, and robustness. PIML has applications across various domains, including fluid dynamics, material science, climate modeling, and biomedical engineering. In fluid dynamics, PIML models demonstrate remarkable accuracy in predicting turbulent flows, significantly advancing simulation capabilities for aerospace and mechanical engineering applications. In material science, PIML accelerates the discovery of novel materials with optimized properties, driving innovation in sectors such as electronics and renewable energy. Climate modeling benefits from the enhanced precision of PIML, crucial for improving weather forecasts and addressing climate change impacts. In biomedical engineering, PIML enhances diagnostic accuracy and treatment planning, contributing to improved patient outcomes and healthcare efficiency.
Date and Time
Location
Hosts
Registration
- Date: 18 Jul 2024
- Time: 10:45 PM UTC to 12:00 AM UTC
-
Add Event to Calendar
- Stockman’s Restaurant, 1175 Pier View Drive
- Idaho Falls, Idaho
- United States 83402-5039
- Contact Event Host
- Co-sponsored by Brew with the Crew Informal Seminar Series - Sponsored by Eastern Idaho IEEE, INL , Idaho ISA
- Starts 16 July 2024 06:00 AM UTC
- Ends 18 July 2024 06:00 AM UTC
- No Admission Charge
Speakers
Bridging Physics and Machine Learning: Real-World Applications
Revanth Mattey graduated with a PhD in Mechanical Engineering from Michigan Technological University and currently works as a post-doc at INL for the Nuclear Fuel Performance and Qualification group. His research focuses on scientific machine learning, computational mechanics, numerical methods, and uncertainty quantification. Ultimately, his goal is to accelerate solutions for problems involving numerical models with machine learning approaches.