On 2024-05-17 I participated in the Chalmers "Energy Network Day" and gave a talk called
"OptiFun: Optimising Fusion with Functional Programming"
The 80-100 participants were from many different research fields and also from companies in the energy area, thus the presentation was at a "popular science" level of detail.
The slides are available here
In the talk I presented some of the ideas and results from a joint project between Computing Science and Physics, funded (with 3M SEK) by the Chalmers "Energy Area of Advance" during 2022-2023. The project had two PIs: Prof Patrik Jansson (that's me;-) and Prof Tünde Fülöp (at the Physics dept.).
We are motivated by two applications: fusion energy experiments and climate impact research and we have published project-related journal papers in both areas. The areas a quite different is the technical details, but the setting is common: real experiments are very expensive (or impossible), thus numerical simulations are needed. These simulations have several input parameters and several objectives (cost, damage, temperature, etc.) we want to minimise.
Some papers cited in the talk:
"OptiFun: Optimising Fusion with Functional Programming"
The 80-100 participants were from many different research fields and also from companies in the energy area, thus the presentation was at a "popular science" level of detail.
The slides are available here
In the talk I presented some of the ideas and results from a joint project between Computing Science and Physics, funded (with 3M SEK) by the Chalmers "Energy Area of Advance" during 2022-2023. The project had two PIs: Prof Patrik Jansson (that's me;-) and Prof Tünde Fülöp (at the Physics dept.).
We are motivated by two applications: fusion energy experiments and climate impact research and we have published project-related journal papers in both areas. The areas a quite different is the technical details, but the setting is common: real experiments are very expensive (or impossible), thus numerical simulations are needed. These simulations have several input parameters and several objectives (cost, damage, temperature, etc.) we want to minimise.
Some papers cited in the talk:
- On the sustainability side:
- The impact of uncertainty on optimal emission policies},
Botta, Jansson, Ionescu.
Earth System Dynamics (2018)
doi:10.5194/esd-9-525-2018 - Responsibility Under Uncertainty: Which Climate Decisions Matter Most?,
Botta et al. incl. Jansson.
Environmental Modeling & Assessment (2023)
doi:10.1007/s10666-022-09867-w
- The impact of uncertainty on optimal emission policies},
- On the fusion physics side:
- Bayesian optimization of massive material injection for disruption mitigation in tokamaks,
Pusztai et al. incl. Jansson & Fülöp.
Journal of Plasma Physics (2023)
doi:10.1017/S0022377823000193 - Fluid and kinetic studies of tokamak disruptions using Bayesian optimization,
Ekmark et al. incl. Jansson & Fülöp,
Journal of Plasma Physics (2024)
arXiv:2402.05843
- Bayesian optimization of massive material injection for disruption mitigation in tokamaks,