Learning from Physics and Data for Design and Condition Monitoring of Engineering Systems

Optimisation Community Online Event

Monday 8th November 2021

08:00 PST (Los Angeles), 11:00 EST (New York)
16:00 GMT (London), 17:00 CET (Berlin)

 

 

Agenda

Event Description

Data-driven methods and machine learning paradigms introduce great opportunities to learn and tailor models of engineering systems and processes suitable for multi-query, time and resource constrained computational tasks. However, data-driven learning techniques mostly require a huge amount of training data which can be very expensive to obtain, gather and process for most applications in science and engineering. In addition, their predictions are commonly hard to characterize in terms of interpretability, robustness and reliability, which are critical to real-life decision making.

This seminar presentedformulations for physics-based machine learning that allow users to cope with small data and offer avenues to hack interpretability issues. Case studies with applications to design and condition monitoring of engineering systems will be reviewed and discussed.

This event was hosted by the NAFEMS Optimisation Working Group. The Optimisation Working Group has formed an online Community to help disseminate best practice and encourage the adoption of optimisation methods and technology. For more information and to get involved go to the Optimisation Community webpage.

You can join the Optimisation Community using the button below:

Join the Optimisation Community