Welcome to the first in a series of Q&As with the ZERO Institute team!
For our first Q&A we’re speaking with Jack Umenberger, a member of the ZERO Institute’s Leadership Team and an Associate Professor in Engineering Science at the University of Oxford.
When and how did you start being involved in the ZERO Institute, what is your background, what did you do before Oxford?
I moved to Oxford in 2023 as a Departmental Lecturer in Engineering Science and started as a Senior Research Fellow in the ZERO Institute in early 2024. Prior to this, I was a postdoc at MIT, working on control and optimization for robotics.
What are your main research areas here, and what are you working at right now?
My research area is (automatic) control and optimisation. ‘Optimization’ is about finding the best solution to a problem, i.e. determining the numerical values of decision variables that achieve the highest (“best”) reward, while satisfying all problem constraints. Roughly speaking, control is optimization in dynamic environments, meaning that decisions made at one time continue to have consequences for decisions made at later times. I am particularly interested in data-driven and machine learning-based approaches to control and optimization. Previously, my main application area was robotics, but now I focus mostly on energy systems. Both domains share similar challenges, such as significant uncertainty. At the moment, I’m working on optimization algorithms to speed up the dispatch of generators in real-time balancing markets (an important problem for managing the increased variability that comes with renewable energy).
What keeps you excited about your work in the ZERO Institute?
The ZERO Institute is relatively new, and what it does and how is still taking shape. In terms of what it can become, I think that connections with industry will be key. A lot of academic research – especially in energy systems, but also more generally — can easily become disconnected from reality. I think impact is best achieved by engaging with the problems that are actually showing up in practice, which means engaging with industry and regulators. (Rapid impact is particularly important if we wish to meet emissions and climate targets). I hope that the ZERO Institute will help build partnerships and connect the university to the real world.
What does success look like for The ZERO Institute?
At least in my area, I would be thrilled if the National Energy System Operator (NESO) would adopt some of our algorithms for control of the grid – particularly algorithms that help control room engineers manage uncertainty in a more systematic, and ultimately more reliable, way. Deployment of new algorithms on critical infrastructure like the grid is challenging; however, NESO has recently invested significantly in modernizing the IT infrastructure that underpins control of the grid (cf. Open Balancing Platform), which should make deployment of new algorithms easier. More broadly, it would be great if the ZERO Institute could play a role in reducing energy prices in GB. High energy prices hurt (almost) everyone, from families to factories. I don’t think there’s one quick fix, and solutions will involve both new technology and new policy (market reform) – but I hope ZERO can play a role.
