CIMantic Graphs Library Documentation

CIMantic Graphs is an open-source library for creating, parsing, and editing CIM power system models using in-memory knowledge graphs to reduce the burden and learning curve associated with using the Common Information Model.

CIM Graph Structure

Key Features

Single API method to obtain data for any CIM class. No more custom database queries.
Single API method to obtain data for EMS node-breaker transmission models, bus-branch planning models, and distribution feeder models.
Single API method for both centralized and distributed architectures.
Multiple database support with no changes to upper-level graph data or API calls. Only need to change host/port specified in ConnectionParameters data object.
Create CIM models from scratch with full object-oriented interface.
Open-source data engineering tool for management of CIM models.
Knowledge graph approach based on semantic understanding of CIM.
Object-oriented data structure with enforcement of CIM Schema.
Data profiles generated directly from Enterprise Architect UML.
Custom profile support using CIMTool.
Direct creation/editing/parsing of CIM XML, JSON-LD.
API support for centralized/distributed transmission + distribution models.

Installation

To install CIMantic Graphs, clone the github repository or use pip install:

pip install cim-graph

Quick Start

import cimgraph.data_profile.cim17v40 as cim
from cimgraph.databases.blazegraph import BlazegraphConnection
from cimgraph.models import FeederModel

# Connect to database
database = BlazegraphConnection()

# Create a feeder model
feeder = database.get_object(mRID="49AD8E07-3BF9-A4E2-CB8F-C3722F837B62")
network = FeederModel(connection=database, container=feeder, distributed=False)

# Query for all line data
network.get_all_edges(cim.ACLineSegment)

# Access graph data
for line in network.graph[cim.ACLineSegment].values():
    print(f"Line: {line.name}, Length: {line.length}")

Documentation Sections

Overview

Get started with CIMantic Graphs, learn about installation, project structure, and how to contribute.

CIM Profiles

Learn how to work with different CIM profiles, build custom profiles, and use CIM objects.

Databases

Connect to various databases including Blazegraph, Neo4j, GraphDB, MySQL, and GridAPPS-D. Also learn how to parse XML and JSON-LD files.

Graph Models

Understand the different graph model types: FeederModel for distribution, NodeBreakerModel for transmission, and BusBranchModel for planning studies.

Utils & Shortcuts

Discover utility functions for file writing, bulk data queries, and automatic mermaid diagram generation.

Support & Contributing

GitHub Repository: PNNL-CIM-Tools/CIM-Graph
PyPI Package: cim-graph
Issue Tracker: GitHub Issues

License

This project is maintained by Pacific Northwest National Laboratory and is available under the terms specified in the repository license.

CIM Graph Logo