Skip to main content

Core data models and I/O for the psforge power system analysis ecosystem. LLM-friendly design for AI-assisted analysis.

Project description

psforge-grid

PyPI version Python versions Tests License: Dual

Hub data model for the psforge power system analysis ecosystem

Core data models and I/O for power system analysis with LLM-friendly design.

Quick Start

pip install psforge-grid
from psforge_grid import System

# Load from any supported format
system = System.from_raw("ieee14.raw")              # PSS/E RAW
system = System.from_matpower("case14.m")            # MATPOWER
system = System.from_pop("WEST10peak.pop")           # CPAT .pop
system = System.from_dyna("model.dyna")              # CPAT .dyna
system = System.from_dss("network.dss")              # OpenDSS
system = System.from_json("ieee14.psfg.json")        # psforge JSON
system = System.from_file("case14.m")                # Auto-detect

# Export to any format
system.to_raw("output.raw")
system.to_matpower("output.m")
system.to_json("output.psfg.json")                   # Human/LLM-friendly
system.to_file("output.m")                           # Auto-detect

# Explore the system
print(f"Buses: {len(system.buses)}, Branches: {len(system.branches)}")
print(system.to_summary())
# Or use the CLI
psforge-grid info ieee14.raw
psforge-grid show case14.m buses -f json

Why psforge-grid?

Feature psforge-grid Others
LLM-friendly output Built-in JSON/summary formats Manual formatting
Educational design Rich docstrings, clear naming Varies
Type hints Complete type annotations Often missing
CLI included Yes, with multiple output formats Usually separate
Multi-format I/O 7 formats: RAW, MATPOWER, CPAT, OpenDSS, JSON Usually single format

Overview

psforge-grid serves as the Hub of the psforge ecosystem, providing:

  • Common data classes (System, Bus, Branch, Generator, GeneratorCost, Load, Shunt)
  • Parsers & Writers for 7 formats (see Supported Formats)
  • Factory pattern: ParserFactory / WriterFactory for format-agnostic I/O
  • Scenario loading: Base case + differential modifications for N-1 / parametric studies
  • OPF data support (GeneratorCost with polynomial and piecewise-linear cost models)
  • Zero-sequence impedance and generator machine parameters for fault/stability analysis

Supported Formats

Format Parse Write Extension Round-trip
PSS/E RAW (v33/v34) Yes Yes (v33) .raw Yes
MATPOWER Yes Yes .m Yes
CPAT .pop (ZIP/XML) Yes Yes .pop Yes
CPAT .dyna (cards) Yes Yes .dyna Yes
OpenDSS Yes Yes .dss Yes
psforge JSON Yes Yes .psfg.json Yes
psforge Scenario Yes Yes .psfg.json -
PSS/E RAW format details

PSS/E RAW

The parser supports core power flow data for AC power flow analysis:

Section v33 v34 Notes
Case Identification Yes Yes Base MVA, system info
Bus Data Yes Yes All bus types (PQ, PV, Slack, Isolated)
Load Data Yes Yes Constant power loads
Fixed Shunt Data Yes Yes Capacitors and reactors
Generator Data Yes Yes P, Q, voltage setpoint, Q limits
Branch Data Yes Yes Transmission lines
Transformer Data Yes Yes Two-winding, magnetizing admittance, is_xfmr flag

Not yet supported: Area/Zone/Owner Data, DC lines, FACTS, Switched Shunts, Three-winding Transformers.

Test data sources:

MATPOWER format details

MATPOWER

Supports MATPOWER .m files, including pglib-opf benchmark cases.

Section Notes
Bus Data (13 columns) All bus types, Vmin/Vmax for OPF
Generator Data (10 columns) Pmin/Pmax, Qmin/Qmax
Branch Data (13 columns) Including angmin/angmax for OPF
Generator Cost Data Polynomial (model=2) and piecewise-linear (model=1)
system = System.from_matpower("pglib_opf_case14_ieee.m")
for cost in system.generator_costs:
    print(cost.to_description())
    # "Generator Cost (polynomial, degree 2): 0.0430 * P^2 + 20.00 * P + 0.00"
CPAT format details (.pop / .dyna)

CPAT

Supports two CPAT formats for IEEJ standard model systems.

.pop format (recommended): CPAT-GUI native project file (ZIP archive + XML).

Feature Status Notes
System data (pnsd) Yes Nodes, branches, generators, loads
Generator machine data Yes G1-G5 card fields (Xd, Xd', Xd'', etc.)
Zero-sequence data Yes Branch and generator zero-sequence impedances

.dyna card format: Legacy Fortran fixed-column (80-character) cards.

Card Type Description
DATA System name, base MVA, frequency
T / X / N Transmission lines / Transformers / Nodes
G1-G5 Generator machine parameters
OpenDSS format details

OpenDSS

Supports OpenDSS .dss script files via opendssdirect.py.

DSSWriter converts per-unit → physical units:

System Element OpenDSS Element
Swing bus New Circuit (Vsource)
Branch (line) New Line
Branch (transformer) New Transformer
Generator New Generator
Load / Shunt New Load / New Capacitor / New Reactor

Fault study mode (write_fault_study()): Y-circuit transformer model with Yg-Delta grounding for zero-sequence analysis.

DSSParser compiles .dss files with OpenDSS and extracts data via API.

psforge JSON format details (.psfg.json)

psforge JSON

Human/LLM-friendly native format with explicit metadata, distinct from pglib-uc JSON.

  • Extension: .psfg.json
  • Metadata: "format": "psforge-grid", "version": "1.0"
  • Field names: snake_case with unit suffixes (_pu, _mw, _kv)
  • Compact output: None fields omitted by default
# Export / Import
system.to_json("ieee14.psfg.json")
system = System.from_json("ieee14.psfg.json")

# Include all fields (with null values)
system.to_json("full.psfg.json", omit_none=False)

Scenario loading (base case + modifications)

Define N-1 contingencies or parametric studies with minimal data:

from psforge_grid.io import load_scenarios, write_scenario

# Define scenarios
write_scenario(
    "contingencies.psfg.json",
    base_case="ieee14.psfg.json",
    scenarios=[
        {
            "name": "N-1_Line_1-5",
            "modifications": [
                {"target": "branches", "match": {"from_bus": 1, "to_bus": 5},
                 "set": {"status": 0}}
            ]
        },
    ],
)

# Load: returns {"base": System, "N-1_Line_1-5": System, ...}
scenarios = load_scenarios("contingencies.psfg.json")

LLM Affinity Design

"Pickaxe in the Gold Rush" - psforge is designed for seamless LLM integration.

Feature Description
Explicit Units Field names include units (voltage_pu, power_mw)
Semantic Status Enum-based status annotations (VoltageStatus.LOW)
Self-Documenting Rich docstrings explaining physical meaning
to_description() Human/LLM-readable output methods
bus = system.get_bus(14)
print(bus.to_description())
# Output: "Bus 14 (LOAD_BUS): 13.8 kV, PQ type"

CLI for LLM Integration

psforge-grid info ieee14.raw              # Table format
psforge-grid info ieee14.raw -f json      # JSON for API/LLM
psforge-grid info ieee14.raw -f summary   # Compact for tokens
psforge-grid show ieee14.raw buses -f json
psforge-grid validate ieee14.raw --strict

Output Formats: table (default), json, summary, csv

See CLAUDE.md for detailed AI development guidelines.

Installation

pip install psforge-grid

# Or install from source
pip install -e .

Development

See docs/development.md for full setup instructions (pre-commit hooks, editor config, etc.).

pip install -e ".[dev]"     # Install dev dependencies
pytest tests/ -v            # Run tests
ruff check src/ tests/      # Lint
mypy src/                   # Type check

psforge Ecosystem (Hub & Spoke Architecture)

                    ┌──────────────────────┐
                    │    psforge-grid      │
                    │   (Hub: Data & I/O)  │
                    └──────────┬───────────┘
                               │
     ┌─────────────┬───────────┼───────────┬─────────────┐
     │             │           │           │             │
┌────▼────┐  ┌────▼────┐ ┌────▼─────┐ ┌───▼────┐  ┌────▼────┐
│ psforge │  │ psforge │ │ psforge- │ │psforge-│  │ psforge │
│  -flow  │  │ -fault  │ │stability │ │schedule│  │ -turbo  │
│ (Power  │  │ (Fault  │ │(Transient│ │ (Unit  │  │  (C++   │
│  Flow)  │  │Analysis)│ │Stability)│ │Commit.)│  │ Engine) │
└─────────┘  └─────────┘ └──────────┘ └────────┘  └─────────┘
Package Description Status
psforge-grid (this) Core data models, parsers, and CLI Active
psforge-flow AC power flow (Newton-Raphson) Active
psforge-fault Short-circuit fault analysis Active
psforge-stability Transient stability analysis Planned
psforge-schedule Unit commitment optimization Planned
psforge-turbo High-performance C++ engine (Phase 2) Frozen

All packages are developed and maintained by Manabe Lab LLC.

Contributing

To maintain the ability to offer commercial licenses and ensure the long-term sustainability of this project, we do not accept external code contributions (Pull Requests) at this time.

We still highly value your feedback — please feel free to open Issues for bug reports and feature suggestions.

License

This software is provided under a dual-licensing model:

  • Individual & Educational Use: MIT License (students, researchers, non-commercial educational purposes)
  • Commercial & Business Use: Requires a Commercial License

See LICENSE for details.


Developed by Manabe Lab LLC

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

psforge_grid-0.6.1.tar.gz (133.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

psforge_grid-0.6.1-py3-none-any.whl (121.8 kB view details)

Uploaded Python 3

File details

Details for the file psforge_grid-0.6.1.tar.gz.

File metadata

  • Download URL: psforge_grid-0.6.1.tar.gz
  • Upload date:
  • Size: 133.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for psforge_grid-0.6.1.tar.gz
Algorithm Hash digest
SHA256 304a56bb8ca133bf111722f71cc1c8ccb7ec1d18c3a7fe791c31e8ce5804f129
MD5 2a8a933cf513c294ea3aee688d98c93f
BLAKE2b-256 c45918a80ccb7472b101a5afd13571b01b80967f9eec5c6fabf126a9193cad88

See more details on using hashes here.

Provenance

The following attestation bundles were made for psforge_grid-0.6.1.tar.gz:

Publisher: publish.yml on manabelab/psforge-grid

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file psforge_grid-0.6.1-py3-none-any.whl.

File metadata

  • Download URL: psforge_grid-0.6.1-py3-none-any.whl
  • Upload date:
  • Size: 121.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for psforge_grid-0.6.1-py3-none-any.whl
Algorithm Hash digest
SHA256 d6c6cbbfb2de9d195980562d0c9788d4f72977fd5b3fbb6b6affd7c3ffe6d946
MD5 d6a251134a5e9e0527c9d0dc553c22df
BLAKE2b-256 77a3aa085c2a9104416ae70dfa2fe8817e23f429c78755fb3f812e740a39b43a

See more details on using hashes here.

Provenance

The following attestation bundles were made for psforge_grid-0.6.1-py3-none-any.whl:

Publisher: publish.yml on manabelab/psforge-grid

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page