Design, optimisation, and research for the energy transition — mostly the Dutch grid, mostly hourly time-series, mostly in Python on top of the open-source PyPSA stack. Independent work runs through Treehouse Energy; investment-grade BESS & grid analytics through Birdview Energy.
Three problems, every project, in some combination:
Concrete work on treehouseenergy.nl — BESS feasibility studies, grid-congestion analyses, hourly time-series simulation, and forward curves for asset valuation.
The pieces below are open-source and most of them have small upstream PRs from me — the Dutch transition is not a problem anyone solves with a closed tool.
The methodology here doesn't come from nowhere. The shortlist of researchers I lean on most for energy-system modelling:
Lead architect of PyPSA and PyPSA-Eur — the framework most of my work runs on. Influential on cost-optimal pathways, sector coupling, and open European energy modelling at TU Berlin.
nworbmot.org · scholarRenewables.ninja, Calliope, and a long line of papers on how weather variability, transmission, and storage interact in high-renewables systems. The reason "your dispatch model is only as honest as its weather year" is the default assumption.
pfenninger.org · renewables.ninjaPrinceton ZERO Lab — capacity expansion, the role of "firm clean" generation, and grid build-out as the binding constraint on the energy transition. Net Zero America and the GenX line of work have shaped how I frame transmission and storage trade-offs in NL studies.
zero-lab.princeton.edu · netzeroamericaCo-author on much of the renewables-availability literature (with Pfenninger), and a steady stream of work on storage economics, hydrogen, and the European power system at Imperial.
imperial profilePyPSA-Eur core developer alongside Tom Brown — high-resolution European modelling, sector coupling, hydrogen networks. Many of the methods I rely on day-to-day were first published as PyPSA-Eur tutorials and notebooks by Fabian.
github.com/fneumTU/e — EV adoption and energy-transition modelling in the Dutch context. A useful sanity check whenever a top-down NL scenario claims something the on-the-ground adoption curve would not support.
tue.nl profileHourly forward power-price curve for NL and neighbouring zones, built bottom-up from fundamentals (fuel forwards, fleet capacity, weather-driven renewables) and calibrated against the cleared forward market. The reference curve I value BESS revenue against.
PyPSA-Eur–based model of the Dutch high-voltage network, calibrated against TenneT data. Used to study congestion, curtailment risk under high-renewables scenarios, and connection-point capacity for new BESS and industrial load.
Simulating day-ahead price formation in 2030 with large-scale BESS and V2G acting and competing — published with code on GitHub and on Google Scholar.
Small but load-bearing PRs to the optimisation stack — linopy, HiGHS, and PyPSA-Eur. Tools I rely on, made faster and less surprising for everyone else solving the same problems.