The data loader supports two JSON formats, detected automatically at runtime. Both can coexist in the same data/ directory.
One file covers all technologies in a domain. Contains a metadata block and a technologies array with flat numeric fields per instance.
Files: data/<category>/<category>_technologies.json
{
"metadata": {
"domain": "generation",
"version": "1.0.0",
"description": "...",
"primary_sources": ["NREL ATB 2023", "IRENA 2023"]
},
"technologies": [
{
"technology_id": "ccgt",
"technology_name": "Combined Cycle Gas Turbine (CCGT)",
"domain": "generation",
"carrier": "natural_gas",
"oeo_class": "http://openenergy-platform.org/ontology/oeo/OEO_00000044",
"description": "...",
"instances": [
{
"instance_id": "ccgt_800mw_current",
"instance_name": "CCGT – 800 MW (Current, 2024)",
"typical_capacity_mw": 800,
"capex_usd_per_kw": 900,
"opex_fixed_usd_per_kw_yr": 20.0,
"opex_var_usd_per_mwh": 3.5,
"efficiency_percent": 58.0,
"lifetime_years": 30,
"co2_emission_factor_operational_g_per_kwh": 202,
"ramping_rate_percent_per_min": 8.0,
"reference_source": "NREL ATB 2023"
}
]
}
]
}
| Catalogue field | Internal field | Conversion |
|---|---|---|
capex_usd_per_kw |
capex_per_kw.value |
direct |
efficiency_percent |
electrical_efficiency.value |
÷ 100 |
typical_capacity_mw |
capacity_kw.value |
× 1000 |
co2_emission_factor_operational_g_per_kwh |
co2_emission_factor.value |
÷ 1000 → tCO₂/MWh |
ramping_rate_percent_per_min |
ramp_up_rate.value + ramp_down_rate.value |
direct |
oeo_class (full URI) |
oeo_uri + oeo_class (last segment) |
split |
One JSON file per technology using nested ParameterValue objects. Detected automatically by the absence of a metadata/technologies root structure.
data/<category>/<category>_technologies.json.technologies array following the schema above.curl -X POST http://localhost:8000/api/v1/debug/reload
GET /api/v1/technologies/{your_new_id}.data/<category>/<tech_id>.json with a nested ParameterValue schema.id field must be a valid UUID.instances.Files in data/timeseries/ are referenced by profile_key on VRE technologies. Metadata is indexed in timeseries_catalogue.json.
{
"profile_id": "de_solar_pv_utility_cf_2019",
"name": "Germany Solar PV Utility CF 2019",
"type": "capacity_factor",
"resolution": "1h",
"location": "DE",
"carrier": "solar",
"year": 2019,
"n_timesteps": 8760,
"source": "...",
"values": [0.0, 0.0, ..., 0.72, ...]
}