2021-12-09 | Blog 007

I'm a bit behind on blog updates, so I'll knock a few out today.

I did a talk for the "Machine Learning in Science" group at the University of Glasgow, a student-run group which brings together those researching, practicing, or otherwise interested in Machine Learning from across different schools at UoG.

My 20-minute talk was on the ways ML can enable the Digital Twin concept for floating offshore wind turbines. The previous half of the session was by Dr Joshua Franz Einsle on “Nanoscale energy critical mineralization: First steps in data driven microanalysis”.

The event was the first one in person for about 2 years. I’m not a great public speaker, but I still think I preferred presenting in person, as opposed to Zoom. Of course, in-person meant we could have a casual discussion after the event finished, which was a highlight for me.

As for self-critique of my own performance:

• I think I should have tried to fit in less material. It’s incredible how fast slides fill up when you want to try explain concepts in full. Future Rory, if you’re reading this, really consider if what you’re adding in is necessary or just a nice-to-know. Edit! Edit! Edit!
• Don’t assume the audience knows what you mean by “sensors on the wind turbine”.
• Speak slower, clearer.

Very happy to have done it, the questions I got at the end indicated the audience were interested. As did the discussion post-event.

Photo below was taken by the event organizers.