Building simulation
IDA ICE work turns building geometry, envelope assumptions and schedules into demand profiles that can feed system-level decisions.
Energy systems modelling
Models that turn heat, power and demand data into engineering decisions.
This track collects the strongest energy-system evidence: IDA ICE building simulation, HOMER Pro island microgrid modelling, SAM and spreadsheet techno-economics, LEAP scenario modelling, Python/PuLP dispatch optimisation, scikit-learn forecasting and distribution-grid studies. PyPSA-NL is retained as a transparent exploratory learning project, while the core track is carried by coursework and project work with clear inputs, tools, assumptions and decision outputs.
6 toolchainsIDA ICE, HOMER Pro, SAM, LEAP, PuLP, scikit-learn
7 core projectsBuildings, district heat, national scenarios, forecasting and grids
3 decision modesTechno-economic, operational dispatch and policy/system scenarios
Evidence map
One track, six modelling behaviours.
Microgrid and autonomy modelling
HOMER Pro work compares renewable supply, storage, grid imports and economics for small energy systems with resilience constraints.
Techno-economic comparison
SAM and spreadsheet models compare CAPEX, OPEX, LCoE, payback and sensitivity across heating and solar alternatives.
Scenario modelling
LEAP work links demand growth, policy assumptions, technology build-out, cost and carbon outcomes in a national energy model.
Optimisation
Python/PuLP dispatch models formalise objective functions and constraints for heat-source allocation, storage and emissions trade-offs.
Forecasting and grid impacts
scikit-learn forecasting and distribution-grid studies connect demand prediction, EV/PV/storage assumptions and operational constraints.
Interactive model stack
Choose the modelling lens.
IDA ICE, HOMER Pro and SAM: from building load to system economics
Hylkysaari and residential heating projects show a complete chain: estimate building demand, compare renewable supply options, size storage or heat pumps, and translate results into payback and design recommendations.
IDA ICEEnvelope, schedules and hourly demand
HOMER ProPV, storage, grid and autonomy economics
SAM / ExcelLCoE, payback and sensitivity
District heating MILP: objective functions make trade-offs explicit
The district heating project moves from base dispatch to storage, future demand, electricity-price coupling and emissions minimisation. It is the cleanest evidence that the modelling work is not only simulation, but decision optimisation.
LEAP: national-scale assumptions made auditable
The Germany analysis links demand projections, eleven generation technologies, policy constraints, merit order, carbon outcomes and social cost. The value is not a single forecast; it is a transparent scenario structure that can be challenged and revised.
Forecasting: model complexity judged against operational value
The heating-demand project compares logistic regression, shallow neural networks and deep neural networks. The important conclusion is engineering judgement: the shallow network is preferred because it generalises better and trains much faster than the DNN for a tiny accuracy penalty.
Distribution-grid studies, with PyPSA kept in the right place
The grid evidence is strongest as supporting systems work: CIGRE-style distribution analysis, N-1 checks, EV/PV/storage impacts and constrained-flow thinking. PyPSA-NL remains visible as an exploratory learning project, not as the primary proof line.
Evidence-page standard
How each project is now meant to be read.
QuestionWhat technical decision does the model answer?
InputsWhat data, assumptions and boundary conditions drive it?
MethodWhich toolchain, equations or optimisation logic was used?
ResultWhat trade-off, KPI or recommendation changed?
LimitsWhere can the model be wrong or incomplete?
Core route cards
Start with these case studies.
IDA ICE + HOMER Pro
Building demand, microgrid, waste heat and economics for Hylkysaari Island.
SAM + heating techno-economics
Solar thermal, PV, heat pump and district-heating comparison for a Stockholm housing cooperative.
Python/PuLP MILP
District heating dispatch with cost, storage, electricity-price and emissions trade-offs.
LEAP 3E scenario model
Germany electricity demand, technology build-out, cost and CO2 scenario comparison.
Heating demand forecasting
Weather and calendar feature modelling for building heating demand prediction.
Distribution-grid study
EV, PV, storage and N-1 impacts on distribution-grid operation.