Standardize or customize? Offshore wind's most expensive design decision is made by judgment.
One common design buys serial production and wastes steel on milder sites. Site-by-site designs do the opposite. Nobody has published a model that prices both sides. I'm building one at NTNU, in the open, over the next three years. The companies that join the study decide which questions it answers first.
— Evgeny Ramenskiy, PhD candidate, NTNU
Twinyard is the public site of a doctoral research project at the Norwegian University of Science and Technology (NTNU). Methods and code are published openly. It is not a commercial product, and there is nothing for sale.
The trade-off looks different from where you stand
Bottom-fixed took fifteen years to converge. Floating doesn't have them.
The monopile became the bottom-fixed standard through fifteen years of costly market iteration. Floating wind has over a hundred concepts and no fifteen years.
So the narrowing has to happen at FEED, deliberately — with numbers on both sides: what over-design costs in steel, and what standardization returns in fabrication.
~24%
of floating CAPEX in substructure & foundation (NREL)
2.8×
floating vs. bottom-fixed substructure cost
100+
competing floating substructure concepts
Honest about timing
No promises of tools that don't exist yet. Today participants get influence and early access. The computational results arrive on the published research schedule.
Now — 2026
openThe needs study
Interviews with yards, developers, consultancies, and ports set the research agenda. Participants steer what gets modeled — and get the free supply chain map today.
2027
in buildThe structural engine
Open-source engine that sizes substructures across site classes and prices the steel penalty of holding a design common. Participants see results first.
2027–2028
plannedThe trade-off model
The steel penalty priced against serial-production savings — the standardize-or-customize curve, run on case studies defined with participants.
2028
plannedThe evaluation
A review panel drawn from participants judges whether the results answer the questions industry actually has. Concrete and hybrid variants extend the scope.
The research is happening either way. Participants decide what it answers.
Start with a 2-minute interest form; the first real step is a 45-minute interview, anonymized by default. Or explore the free supply chain map and the research itself — or challenge the premise. That's useful too.