What Denmark can build

The generators in DK1 and DK2, with the costs the model weighs them by. For each expandable technology the model chooses the capacity, up to the upper limit; the others are fixed. Everything outside Denmark is fixed too.

As stored in the network files. For expandable technologies the file also holds a p_nom, which the capacity-expansion step does not use: it chooses the capacity itself. Capital cost is the annualised fixed cost the expansion step weighs (capital_cost); fixed O&M (fom_cost) and variable cost (marginal_cost) as given. For dispatchable plant the variable cost rises with output (marginal_cost_quadratic), a fleet of units of varying cost. Mean availability is the average of the hourly availability profile; wind and solar must produce exactly what the weather allows. Model names are DNA-PyPSA’s; technology names follow the report.

Weather and demand, hour by hour

Wind, sun and temperature-driven demand all come from one weather year, 1991, chosen because its price volatility and capture rates are close to the long-run average of the 33 years in the dataset. Demand peaks and troughs therefore line up with the wind and sun the way they did that year.

Flexibility in Danish demand

Not all demand is fixed. Flexible demand has a buffer: it can buy electricity ahead, in cheaper hours, and hold it until it is needed, but it cannot postpone what it needs. Price-sensitive demand is only served while the price is below what it is willing to pay, and that willingness falls the more of it is served.

Connections to the neighbours

Transmission between zones is fixed, at capacities based on ENTSO-E plans; the model does not expand it. Each border is two one-way links that lose 5 % of what they carry.

The system around Denmark

The other 15 zones have fixed fleets, set by a separate optimisation of the European power system with the Balanced scenario as reference. They give Denmark realistic prices and trade at its borders without being optimised alongside it, since national energy policies are in practice made independently.

How the run is set up, and how to repeat it

# get the model and its inputs
git clone https://github.com/QuantifiedCarbon/DNA-PyPSA.git
cd DNA-PyPSA && pip install .

# solve the Balanced scenario
python scripts/run_full_scenarios.py --scenario reference