One file. The whole grid.
For ops and planning.
The .rpf is a modern, all-in-one “digital briefcase” that holds almost everything a utility needs to run both real-time operations and planning studies — in a single fast, reliable, self-describing package.
The .rpf file is not a plain-text file you can open in Notepad like a legacy PSS/E .raw — it’s a compact binary format. But it is deliberately built to be practical for humans.
Any engineer can instantly view, search, or export the data with free tools (Raptrix Studio, Python, or simple CLI viewers), and it comes with plain-English field names and metadata so you don’t need a PhD to understand what’s inside.
Even though it carries way more useful information than legacy formats, it is still dramatically cleaner, smaller, and faster to read and write. It is versioned so old and new files work together (forward / backward compatible), and it has built-in integrity checks (a deterministic fingerprint) that make it easy to prove the file hasn’t been tampered with.
In short: rich, modern, and practical — without the bloat or slowness of the old ways.
Millisecond-scale, on a Texas 7k-bus reference case.
Arrow columnar + zero-copy beats text parsing of legacy .raw/.epc by an order of magnitude on large cases. Internal-benchmark numbers below are on a Texas 7k-bus reference case — not a vendor comparison.
One container. The whole grid.
- busesFull network topology
- branchesLines, transformers, FACTS, PSTs
- generatorsLimits, ramp rates, PV/PQ
- shunts / loadsDiscrete steps, control groups
- buses_solvedV, angle per bus
- branches_solvedMW / MVAR flows
- q_injectionsQ-limit enforcement results
- contingenciesNative definition + outcomes
- interfacesTransfer limits + violations
- computational_load_profilesNERC L3 data centers, ramps, guardrails, reductions
- areas / zones / ownersDiscrete asset ownership
- provenanceWho ran it, when, solver iterations, convergence
Self-describing schema embedded in every file. Forward + backward compatible.
Deterministic metadata.fingerprint over the canonical data.
Same input → same fingerprint. Easy to sign or verify with standard tools.
18 root tables. One self-describing container.
Buses, branches, generators, shunts, transformers, loads.
Separate buses_solved, branches_solved, etc. with real V, angles, MW/MVAR flows, Q injections.
A case_mode field explicitly distinguishes planning vs solved / operational cases.
Native tables for contingencies, interfaces, transfer limits, and violation results.
Seasonal envelopes, buildout ramps, voltage guardrails, sudden ramp/drop behavior, load reduction schemes — the NERC Level 3 data path.
IBR-specific metadata and strict Q-limit enforcement outcomes.
Discrete asset ownership metadata.
Who ran it, when, solver iterations, convergence details, Q-limit violations flagged, PV→PQ switches.
Designed for the way utilities actually work.
Zero fidelity loss between Sentinel and Forge or any downstream planning tool.
Optional one-line diagram layout data stored in the file itself.
Via the public raptrix-cim-rs converter.
Exact same input → exact same fingerprint.
Designed for 5–15 min real-time cycles and massive planning loops on laptop-grade hardware.
Pure physics and operational data. Layer market tools on top via CIM or other exports.
Covers standard power-flow and N-1 / N-2 modeling for computational load compliance.
The core .rpf format plus the computational_load_profiles table and contingency outcome fields cover seasonal envelopes, buildout ramps, voltage guardrails, sudden ramp/drop behavior, load reduction schemes, IBR interaction, solved physics, and recovery assessment. A lightweight JSON sidecar is reserved only for supplementary telemetry, custom instrumentation, or narrative that doesn’t belong in the deterministic Arrow contract — keeping .rpf fast, self-describing, and schema-stable while still giving full regulatory flexibility.
The parsers and Studio are planned open-source. MPL 2.0.
Open .rpf interchange ensures transparent data handling and future-proof workflows. Today the .rpf format is feeding Sentinel and Forge; tomorrow it can carry your interconnection workflows, academic syllabi, and downstream SCUC/SCED engines.
Want .rpf in your workflow?
Pilot Sentinel on a recent SE case, or talk to us about integrating .rpf into your existing planning and interconnection workflows.
info@raptrixpower.com · Founder direct: matthew.musto@raptrixpower.com · github.com/RaptrixPowerFlow