Energy Modeling and Simulation using Python
Budget: €8 – €30 EUR
I need a reproducible Python-based model, built with framework, that optimises energy consumption in residential buildings at a state-wide scale. The focus is squarely on heating and cooling systems; lighting, appliances or other end-uses can remain outside the scope for now.
What I will provide
• Hourly electricity-use data for representative homes across the state
• Local weather profiles and tariff structures
• Any policy constraints that must be reflected in the optimisation
What I need back
1. A clean, well-commented oemof / solph model (Python scripts or a Jupyter notebook) that:
• Ingests the data sets above
• Performs cost-optimised scheduling of heating and cooling loads
• Outputs key metrics such as total energy saved, peak reduction and cost impacts
2. A short read-me explaining the input format, assumptions, and steps to reproduce the results.
3. A concise results brief with plots (matplotlib or plotly) summarising the optimisation outcomes.
Acceptance criteria
• The code runs end-to-end in a fresh virtual environment using pip-installable packages only.
• Results reproduce within ±2 % of your sample output.
• All functions and classes are commented so I can extend the model later (for example, to add lighting or renewable integration).
Please keep the solution lightweight—pandas, numpy and similar libraries are fine, but no heavyweight, unnecessary dependencies.
What I will provide
• Hourly electricity-use data for representative homes across the state
• Local weather profiles and tariff structures
• Any policy constraints that must be reflected in the optimisation
What I need back
1. A clean, well-commented oemof / solph model (Python scripts or a Jupyter notebook) that:
• Ingests the data sets above
• Performs cost-optimised scheduling of heating and cooling loads
• Outputs key metrics such as total energy saved, peak reduction and cost impacts
2. A short read-me explaining the input format, assumptions, and steps to reproduce the results.
3. A concise results brief with plots (matplotlib or plotly) summarising the optimisation outcomes.
Acceptance criteria
• The code runs end-to-end in a fresh virtual environment using pip-installable packages only.
• Results reproduce within ±2 % of your sample output.
• All functions and classes are commented so I can extend the model later (for example, to add lighting or renewable integration).
Please keep the solution lightweight—pandas, numpy and similar libraries are fine, but no heavyweight, unnecessary dependencies.