SQLite-backed Python API for reading, building, and writing PLEXOS XML models
plexosdb converts PLEXOS XML input files into an in-memory SQLite database, giving you a fast, typed Python API to query, create, and modify power system models programmatically, then write them back to XML.
pip install plexosdbOr with uv:
uv add plexosdbPython version support: 3.11, 3.12, 3.13, 3.14
from plexosdb import PlexosDB, ClassEnum, CollectionEnum
# Load a PLEXOS XML file into an in-memory SQLite database
db = PlexosDB.from_xml("model.xml")
# Query existing objects
generators = db.get_object_names(ClassEnum.Generator)
# Add new objects
db.add_object(ClassEnum.Generator, name="SolarPV_01", category="Renewables")
db.add_object(ClassEnum.Node, name="Bus_1")
# Create memberships between objects
db.add_membership(
CollectionEnum.GeneratorNodes,
parent_class=ClassEnum.Generator,
parent_name="SolarPV_01",
child_class=ClassEnum.Node,
child_name="Bus_1",
)
# Export the modified model back to XML
db.to_xml("modified_model.xml")When creating a brand-new database (not importing XML), you can preload the matching PLEXOS master template by version:
from plexosdb import PlexosDB
db = PlexosDB()
db.create_schema(version=10)Supported versions: 9, 10, 11, 12.
Accepted inputs include integers, strings, and tuples. For example:
db.create_schema(version="v11.0R4")
db.create_schema(version=(12, 0, 3))from plexosdb import PlexosSolution
import pandas as pd
PLEXOS_SOLUTION = "/path/to/solution.zip"
sol = PlexosSolution.from_zip(PLEXOS_SOLUTION)
sol.to_sqlite("output.sqlite", if_exists="replace")
table = "ST__Interval__Regions__Fixed_Load"
sol.materialize_table(table, schema="report")
df_table = pd.read_sql_query(f'SELECT * FROM report."{table}"', sol.connection)This pattern is useful when you only need a few report/data tables from a large solution ZIP, because it materializes tables on demand.
Full documentation is available at natlabrockies.github.io/plexosdb.
For related previous/current work on querying PLEXOS outputs with DuckDB, see plexos2duckdb.
plexosdb uses uv for dependency management.
git clone https://github.com/NatLabRockies/plexosdb.git
cd plexosdb
uv sync --all-groupsInstall the git hooks so your code is checked before making commits:
uv run prek installRun the test suite:
uv run pytestThis software is released under a BSD-3-Clause License.
This software was developed under software record SWR-24-90 at the National Renewable Energy Laboratory (NREL).
PLEXOS is a registered trademark of Energy Exemplar Pty Ltd. Energy Exemplar Pty Ltd. has no affiliation to or participation in this software. Reference herein to any specific commercial products, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or Alliance for Sustainable Energy, LLC ("Alliance"). The views and opinions of authors expressed in the available or referenced documents do not necessarily state or reflect those of the United States Government or Alliance.