Storage in OSeMOSYS

Long-term energy models often simulate only a handful of "representative days" instead of every hour of the year, which keeps them fast but makes energy storage hard to capture. This site accompanies a study that compares four ways of modelling storage in OSeMOSYS under that simplification, and shares the code to rerun it. The four cards below follow the natural reading order: the idea, how the models are run, what the study found, and the code behind it.

01 Concept Why storage is hard to model with representative days, how the study is put together, and the four formulations. 02 Modelling workflow Installation, running the pipeline and setting up new scenarios. 03 Results Reproduce the published study, read the output files and analyse them. 04 API What each module in src does and how it can be reused in other work.

New to the study? Start with how the study is put together: an interactive map of the test system, model, solve and recommendations.

At a glance

git clone https://github.com/DeltaE/storage-in-OSeMOSYS.git
cd storage-in-OSeMOSYS
make setup && conda activate storage_osemosys
python workflow/main.py

See Quickstart for what this does and where the results go.

Citation

Under review

Borba, B., Md Eliasinul Islam, & Niet, T. Storage in Long-Term Energy Planning Models with Temporal Aggregation: Best Practices and Algorithms. Environmental Research Communications (2026).

Developed at the Delta E+ Research Lab, Simon Fraser University · MIT License