Europe is trying to do a lot at once. Clean up the grid, keep the lights on, keep bills manageable, and get hydrogen, storage, and electric everything up and running, all at the same time. It's a lot to juggle, and the decisions being made today will shape how energy gets produced and used for decades.
That was the backdrop for Xcelerate London, where the people who plan, build, and run Europe's energy systems got together to compare notes. After a dozen customer case studies, a few big ideas kept coming up.
Here are seven of them.
AI came up a lot, and the excitement was real. From PLEXOS® Intelligence, dispatch optimization, multi-market analysis, and constraint checking at a scale no team could match by hand. But the jump from insight to decision keeps landing back with a person, and for good reason. Getting to a credible answer takes a back and forth where the model produces something, an expert checks whether it makes engineering sense, assumptions get refined, and the model runs again. The best version of this future isn't AI replacing people. It's AI clearing away the busy work so people can spend their time on the work that actually need a human judgment call.
Heating, transport, and industry are all racing to plug into the same clean power supply, and they're not always racing on the same clock. A country can hit its own power sector targets and still get caught out, because heat pumps, electric vehicles, and industrial electrification don't wait politely for their turn. When those sectors are modeled in isolation from the grid, planners end up underbuilding for demand that shows up faster than expected, or overbuilding capacity that never gets used because the sector it was meant to serve is running behind schedule. The lesson isn't just "don't ignore other sectors." It's that the timing has to be modeled together, or the numbers on each side stop meaning anything.
Cross-border links that pull double duty (connecting two countries and hooking up offshore wind at the same time) are, on paper, the more efficient option. In practice, most offshore wind capacity still gets built radial-only, because developers plan around their home market first and don't build to the spec that would let it plug into a shared connection later. There's a similar mismatch between neighbors. One country ramps up renewables assuming it can export the surplus, only for its neighbor to ramp up its own capacity on a different timeline and close that door. Coordinating the technology is the easy part. Coordinating the timing is where it actually gets hard.
Hydrogen is one of those topics where everyone agrees it matters, but nobody agrees on the details yet. Some national strategies set firm production targets and then go quiet on infrastructure, effectively leaving the pipelines and storage to be sorted out later by whoever ends up using them. That gap between production ambition and infrastructure commitment is where the real hydrogen debate is happening. As cheap hydrogen becomes available from lower-cost regions, importing it can beat producing it at home, which only strengthens the case for building the infrastructure to move it around rather than waiting for the strategy documents to catch up.
Storage that can hold power for days, not just hours, is quickly moving from an interesting idea to a real business decision. The catch is that a multi-day asset doesn't behave like a battery with the duration turned up. It needs a longer optimization horizon, a much closer read on physical engineering constraints, and the ability to hold onto flexibility for when it's actually worth cashing in. Model it like a stretched-out battery instead, and it's easy to quietly erode a big share of what the asset is worth before it's even built. This isn't a future problem anymore. It's a today problem, with real money on the line.
Some of the most interesting ideas at Xcelerate had nothing to do with power plants at all. Data centers are pairing with heat pumps to send waste heat straight into district heating networks instead of losing it. Surplus renewable power that would otherwise be curtailed is getting converted and stored as heat for winter, cutting reliance on the fossil fuel boilers currently doing that job. None of this shows up as a saving if you model the data center, the heat network, and the grid separately. It only shows up once you model them as one system.
Detailed siting work is happening right now, breaking a country's land and sea area down into thousands of small parcels and scores each one for the specific technology it could host. Salt formations suitable for hydrogen storage, sites with the cooling water access certain generation types need, land that communities will and won't accept new infrastructure on. Layer environmental and social constraints on top of that and figuring out where things go is turning into as big a job as figuring out what to build in the first place.
None of this was theoretical. These are real decisions, with real budgets and real infrastructure attached to them, being made by the people who keep Europe's lights on.
Energy Exemplar's PLEXOS® platform is built to help with exactly this kind of complexity. Whether that's syncing sector timing across heat, transport, and power, valuing a storage asset properly, or figuring out where a hydrogen hub actually makes sense to build, the answers only hold up if the modeling captures how tightly these pieces are wired together.
Europe's energy transition isn't one simple story. It's a lot of moving, interdependent parts, and the teams who get ahead are the ones whose planning can keep up with all of them.