microsoft/GridSFM_US_power_grid

GridSFM US Power Grid Dataset

Overview

GridSFM US Power Grid Dataset is a set of geographically grounded, electrically coherent power-system network derived entirely from publicly available data. It was developed to support AC optimal power flow (AC-OPF) analysis, enabling physics-based study of congestion, capacity, and demand sitting without restricted data.

A detailed discussion of GridSFM US Power Grid Dataset, including how it was developed and evaluated, can be found in our paper at: https://arxiv.org/abs/2605.04289.

Intended uses

The GridSFM US Power Grid Dataset is intended to support a broad range of physics-based research questions on the U.S. transmission network, covering 48 states and multi-state interconnections with realistic geographic structure, including transmission expansion potential, targeted line upgrades, and placement of large loads.

Out-of-scope uses

GridSFM US Power Grid Dataset is not well suited for detailed operational or market-critical decision making, including real-time dispatch, contingency analysis, or regulatory planning that requires exact system parameters.

There are few or no instances of measured electrical parameters, complete multi-circuit topology, detailed protection models, or operational control parameters in this dataset. As a result, GridSFM US Power Grid Dataset should not be used for safety-critical, financial, or regulatory decisions that depend on precise modeling of the real transmission grid.

We do not recommend using GridSFM US Power Grid Dataset in commercial or real-world applications without further testing and development. It is being released for research purposes.

We do not recommend using GridSFM US Power Grid Dataset in the context of high-risk decision making (e.g. in law enforcement, legal, finance, or healthcare).

Citation

If you use this dataset, please cite:

@article{britto2026powergrid,
  title   = {Building Power Grid Models from Open Data: A Complete Pipeline from OpenStreetMap to Optimal Power Flow},
  author  = {Britto, Andrea and Spina, Thiago and Yang, Weiwei and Fowers, Spencer and Zhang, Baosen and White, Chris},
  year    = {2026},
  url     = {https://arxiv.org/abs/2605.04289}
  note    = {Microsoft Research}
}

Dataset Details

Dataset Contents

GridSFM US Power Grid Dataset consists of 54 instances of OPF-ready transmission network models (48 contiguous U.S. states and 6 multi-state regions), derived entirely from open data.

Each instance includes bus-branch topology, estimated electrical parameters (line impedances, transformer characteristics), generator attributes (capacity, fuel type, cost functions), hourly demand allocation, DC warm-start solution (voltage angles and dispatch), and synthetic reactive compensation shunts for AC-OPF feasibility.

Each instance is associated with a label describing the geographic scope (state or multi-state region) and operating condition (peak 16h or off-peak 04h demand snapshot).

The data was generated between 2025 and 2026.

GridSFM US Power Grid Dataset does not contain links to external data sources. Links are to publicly available datasets used during generation (e.g., OpenStreetMap, U.S. EIA, U.S. Census, HIFLD), which are referenced for provenance but not dynamically queried at runtime.

Data Creation & Processing

GridSFM US Power Grid Dataset was created from scratch, using publicly available open data sources, rather than adapting any existing power grid datasets.

The existing data that was used to create GridSFM US Power Grid Dataset consisted of geospatial descriptions of power infrastructure (transmission lines, substations, and generators), generator metadata (capacity, fuel type, heat rates), hourly demand measurements at the balancing-authority level, population-based geographic data, and boundary definitions for balancing authorities.

The existing data that was used to create GridSFM US Power Grid Dataset was originally collected by external public data providers, including OpenStreetMap contributors (crowdsourced mapping), the U.S. Energy Information Administration (EIA), the U.S. Census Bureau, and U.S. government infrastructure datasets such as HIFLD.

GridSFM US Power Grid Dataset was created by transforming these heterogeneous data sources into transmission network models through a multi-stage pipeline, including data extraction, topology reconstruction, parameter estimation, demand allocation, and optimal power flow (OPF) solving with progressive relaxation.

Dataset creation was carried out by members of the Microsoft Research Catalyst Lab team.

GridSFM US Power Grid Dataset includes data crawled from the web. Specifically, OpenStreetMap data (power infrastructure features) was programmatically retrieved via local copy using the Overpass API, which serves publicly available, user-contributed geographic data.

People & Identifiers

The GridSFM US Power Grid Dataset is not related to humans in any way and thus does not include any information that could be used to identify a person.

Sensitive or harmful content

GridSFM US Power Grid Dataset contains only power systems information and thus no sensitive or harmful content.

Other processing

Duplicate/redundant information was automatically removed using software, as part of the topology reconstruction and data preprocessing pipeline.

The data was labeled with metadata describing the modeled region (state or multi-state region), operating condition (peak or off-peak hour), and solver outputs (e.g., DC/AC-OPF results and relaxation levels). The labeling was performed automatically using software, based on deterministic naming conventions and outputs from the OPF pipeline.

How to get started

To begin using GridSFM US Power Grid Dataset, users can download and load the dataset directly from Hugging Face or use the official loader from the GridSFM repository:

• microsoft/GridSFM_US_power_grid · Datasets at Hugging Face
• microsoft/GridSFM: Small Foundation Models for the Power Grid

See section Dataset Download, Usage, and File Specification for detailed code examples.

Validation

To assess how effective GridSFM US Power Grid Dataset would be at its intended purpose, our team looked for physical plausibility, solver feasibility, and consistency with real-world power system statistics.

Specifically, we:

• Evaluated DC-OPF and AC-OPF convergence rates across all 48 states and multi-state regions
• Measured dispatch costs, system losses, and generator utilization, comparing them to expected ranges in real-world systems
• Assessed model robustness under different relaxation levels, using convergence behavior as a proxy for data quality
• Verified scaling consistency across geographic regions, from small states to continent-scale interconnections

A detailed discussion of our validation methods and results can be found in our paper at: https://arxiv.org/abs/2605.04289

Limitations

GridSFM US Power Grid Dataset was developed for research and experimental purposes. Further testing and validation are needed before considering its application in commercial or real-world scenarios.

GridSFM US Power Grid Dataset consists of English language instances only.

GridSFM US Power Grid Dataset contains approximate and inferred data, including estimation noise in electrical parameters (e.g., line impedances, thermal limits), incomplete topology reconstruction (e.g., missing parallel circuits), and heuristic demand allocation.

GridSFM US Power Grid Dataset is missing utility-grade measurements and operational data, including exact network topology, measured electrical parameters, protection settings, dynamic system behavior, and time-series operational constraints.

There are few or no instances of detailed distribution-level networks, low-voltage infrastructure, or precise multi-circuit transmission representations in the dataset. As a result, GridSFM US Power Grid Dataset should not be used for high-fidelity operational studies, protection analysis, or real-time decision-making.

The ability to access external links in the dataset is beyond the control of the research team.

GridSFM US Power Grid Dataset should not be used in highly regulated domains where inaccurate or incomplete outputs could suggest actions that lead to injury or negatively impact an individual's legal, financial, or life opportunities.

Best Practices

We recommend splitting the data into train/validation/test splits based on geographic regions (e.g., by state or multi-state region) or operating conditions (peak vs off-peak), depending on the intended use case.

It is the user’s responsibility to ensure that the use of GridSFM US Power Grid Dataset complies with relevant data protection regulations and organizational guidelines.

License

MIT License

Nothing disclosed here, including the Out of Scope Uses section, should be interpreted as or deemed a restriction or modification to the license the code is released under.

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.

Contact

This research was conducted by members of Microsoft Research. We welcome feedback and collaboration from our audience. If you have suggestions, questions, or observe unexpected/problematic data in our dataset, please contact us at gridsfm@microsoft.com.

If the team receives reports of undesired content or identifies issues independently, we will update this repository with appropriate mitigations.

---

Dataset Content, Quick Start, and Specification

OPF-ready transmission network models for all 48 contiguous U.S. states and 6 multi-state regions, derived entirely from open data (OpenStreetMap + U.S. EIA + U.S. Census).

Each model is a self-contained JSON file compatible with PowerModels.jl and MATPOWER-format tools. Models include bus-branch topology, line impedances, generator costs, hourly load allocation, DC warm-start voltage angles, and reactive compensation shunts.

Tools & Viewer: The Python loader (gridsfm_pg_loader.py) and the interactive Data Viewer are available in the GridSFM repository.

Coverage

48 states — all contiguous U.S. states (AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY).

6 multi-state regions:

Region	States	Buses (approx.)
new_england	CT, MA, ME, NH, RI, VT	~640
pacific_nw	OR, WA	~1,100
desert_sw	AZ, NV, UT	~1,300
western	AZ, CA, CO, ID, MT, NM, NV, OR, UT, WA, WY	~5,100
eastern	AL, AR, CT, DE, FL, GA, IA, IL, IN, KS, KY, LA, MA, MD, ME, MI, MN, MO, MS, NC, ND, NE, NH, NJ, NY, OH, OK, PA, RI, SC, SD, TN, VA, VT, WI, WV	~21,700
pjm	DE, IL, IN, KY, MD, MI, NC, NJ, OH, PA, TN, VA, WV	~7,800

Two operating hours per model:

• 16h — peak demand (4:00 PM, July 15 2024)
• 04h — off-peak demand (4:00 AM, July 15 2024)

File Structure

16h/
  alabama_model.json                 # Full model (topology + parameters + demand + shunts + DC warm-start)
  alabama_dc_results.json            # DC-OPF solution
  alabama_ac_results.json            # AC-OPF solution
  ...
  western_model.json                 # Multi-state region model
  western_dc_results.json
  western_ac_results.json
04h/
  ...                                # Same structure, off-peak hour

Quick Start

Download from HuggingFace

pip install huggingface_hub

Load a single model (no extra dependencies):

from huggingface_hub import hf_hub_download
import json

path = hf_hub_download(
    repo_id="microsoft/GridSFM_US_power_grid",
    filename="16h/texas_model.json",
    repo_type="dataset",
)
with open(path) as f:
    model = json.load(f)

print(f"Buses: {len(model['bus'])}")
print(f"Branches: {len(model['branch'])}")
print(f"Generators: {len(model['gen'])}")
print(f"Loads: {len(model['load'])}")
print(f"Total load: {sum(l['pd'] for l in model['load'].values()) * model['baseMVA']:.0f} MW")

Using the GridSFM loader (from github.com/microsoft/GridSFM/power_grid):

from gridsfm_pg_loader import GridSFM_PG_Loader

# With export_dir (optional): the entire dataset is automatically downloaded
# to this directory on init. Without it, files are stored in HuggingFace's cache.
loader = GridSFM_PG_Loader(
    "microsoft/GridSFM_US_power_grid",
    export_dir="./gridsfm_data",  # optional; pre-fetches everything here
)

# To skip the automatic download, use pre_fetch_all=False (lazy download on access)
# loader = GridSFM_PG_Loader(
#     "microsoft/GridSFM_US_power_grid",
#     export_dir="./gridsfm_data",
#     pre_fetch_all=False,
# )

# Case-insensitive; state abbreviations work too
model = loader.load_model("TX", hour="16h")
ac    = loader.load_ac_results("texas", hour="16h")
dc    = loader.load_dc_results("Texas", hour="16h")

# Export a single file to a specific path
loader.export_file("TX", "model", hour="16h", dest="./my_models/texas.json")

# Save a loaded (or modified) model dict back to JSON
loader.save_json(model, "./my_models/texas_modified.json")

# Discover what's available (fetched from dataset_metadata.json)
loader.list_regions()        # all 54 regions + states
loader.list_abbreviations()  # {"AL": "alabama", "TX": "texas", ...}
loader.list_hours()          # ["04h", "16h"]
loader.list_file_types()     # ["model", "ac_results", "dc_results"]

# Export the entire dataset to a local directory at any time
loader.export_all("./gridsfm_data")

Download the entire dataset (~230 MB):

hf download --repo-type dataset microsoft/GridSFM_US_power_grid --local-dir ./gridsfm_data

Download a subset (one hour only):

hf download --repo-type dataset microsoft/GridSFM_US_power_grid --include "16h/*" --local-dir ./gridsfm_data

Load a model from local files

import json

with open("16h/texas_model.json") as f:
    model = json.load(f)

print(f"Buses: {len(model['bus'])}")
print(f"Branches: {len(model['branch'])}")
print(f"Generators: {len(model['gen'])}")
print(f"Loads: {len(model['load'])}")
print(f"Shunts: {len(model['shunt'])}")
print(f"HVDC lines: {len(model['dcline'])}")
print(f"Total load: {sum(l['pd'] for l in model['load'].values()) * model['baseMVA']:.0f} MW")

Run OPF with PowerModels.jl

using PowerModels, Ipopt

data = PowerModels.parse_file("16h/texas_model.json")
result = solve_ac_opf(data, Ipopt.Optimizer)
println("Objective: \$(result[\"objective\"])")
println("Status: \$(result[\"termination_status\"])")

Model File (*_model.json)

The primary artifact. A single JSON containing everything needed to run optimal power flow.

Top-Level Metadata

Field	Type	Description
name	string	Model name (e.g., "delaware")
baseMVA	float	System base power (always 100.0)
per_unit	bool	Always true — all values are in per-unit
version	string	Format version
source_type	string	"matpower" — format compatibility marker
balancing_authority	string	Primary BA serving this state (e.g., "PJM")
demand_source	string	EIA data source and allocation fraction
dispatch_method	string	Generator dispatch method ("merit_order" or "proportional")
load_allocation_method	string	How load was distributed to buses ("census" or "per_ba_census")
ba_coverage_pct	float	Percentage of state capacity covered by detected BAs
is_multi_state	bool	Whether this is a multi-state region model
target_datetime	string	ISO 8601 timestamp for demand snapshot

storage and switch are present but always empty (required by PowerModels.jl parser).

bus — Transmission Buses

Keyed by string ID (non-sequential). One bus per voltage level per substation.

Field	Type	Description
bus_i	int	Bus number
bus_type	int	1 = PQ, 2 = PV (generator), 3 = slack (reference)
index	int	Same as bus_i
name	string	Substation name from OSM
area	int	Network area
zone	int	Network zone
base_kv	float	Nominal voltage (kV)
lat	float	Latitude (WGS84)
lon	float	Longitude (WGS84)
vm	float	Voltage magnitude (p.u.) — initialized to 1.0
va	float	Voltage angle (radians) — from DC-OPF warm-start
vmax	float	Upper voltage limit (p.u.)
vmin	float	Lower voltage limit (p.u.)
pd	float	Always 0.0 (loads are in the load section)
qd	float	Always 0.0

gen — Generators

Field	Type	Description
index	int	Generator number
gen_bus	int	Bus this generator is connected to
gen_status	int	1 = online, 0 = offline
name	string	Generator/plant name from OSM
fuel_type	string	Standardized fuel type ("gas", "nuclear", "solar", "wind", "coal", "hydro", etc.)
pg	float	Active power output (p.u.) — from DC warm-start dispatch
pmax	float	Maximum active power (p.u.)
pmin	float	Minimum active power (p.u.)
qg	float	Reactive power output (p.u.)
qmax	float	Maximum reactive power (p.u.)
qmin	float	Minimum reactive power (p.u.)
vg	float	Voltage setpoint (p.u.)
mbase	float	Machine base (MVA)
model	int	Cost model type (2 = polynomial)
ncost	int	Number of cost coefficients
cost	list	Cost polynomial [c2, c1, c0] where total_cost = c2·pg² + c1·pg + c0 (p.u.)
apf	float	Area participation factor
startup	float	Startup cost ($)
shutdown	float	Shutdown cost ($)
ramp_10	float	10-minute ramp rate (p.u.)
ramp_30	float	30-minute ramp rate (p.u.)
ramp_agc	float	AGC ramp rate (p.u.)
ramp_q	float	Reactive ramp rate (p.u.)
startup_time	float	Startup time (hours)
min_up_time	float	Minimum up time (hours)
min_down_time	float	Minimum down time (hours)

EIA-matched generators also have:

Field	Type	Description
fuel_type_eia	string	Raw EIA fuel code ("NG", "SUN", "NUC", etc.)
prime_mover	string	EIA prime mover code ("CT", "PV", "ST", etc.)
eia_plant_id	string	EIA plant ID
eia_generator_id	string	EIA generator ID
eia_match_score	float	Match confidence (0–1)
eia_match_distance_km	float	Distance from OSM location to EIA plant (km)
ref_us_eia	string	EIA reference ID
pmax_eia	float	EIA nameplate capacity (p.u.)
heat_rate_eia	float	EIA heat rate (BTU/kWh) — thermal generators only
capacity_factor	float	Hourly capacity factor (solar/wind derating)
pmax_nameplate	float	Nameplate pmax before capacity factor derating (p.u.)
qmax_nameplate	float	Nameplate qmax before any adjustments (p.u.)
qmin_nameplate	float	Nameplate qmin before any adjustments (p.u.)

Injected generators (from distant EIA plants with no OSM match) also have:

Field	Type	Description
eia_injected	bool	Always true — generator was injected, not matched to OSM

branch — Transmission Lines and Transformers

Field	Type	Description
index	int	Branch number
f_bus	int	From bus
t_bus	int	To bus
br_r	float	Series resistance (p.u.)
br_x	float	Series reactance (p.u.)
b_fr	float	From-side shunt susceptance (p.u.)
b_to	float	To-side shunt susceptance (p.u.)
g_fr	float	From-side shunt conductance (p.u.)
g_to	float	To-side shunt conductance (p.u.)
br_status	int	1 = in service
rate_a	float	Long-term thermal rating (p.u.)
rate_b	float	Short-term rating (p.u.)
rate_c	float	Emergency rating (p.u.)
angmin	float	Minimum angle difference (radians)
angmax	float	Maximum angle difference (radians)
tap	float	Tap ratio (1.0 for lines; off-nominal for transformers)
tap_min	float	Minimum tap ratio
tap_max	float	Maximum tap ratio
shift	float	Phase shift angle (radians)
transformer	bool	true if this is a transformer
circuit_key	string	Internal circuit identifier
length_km	float	Line length in km (0.0 for transformers)

Transformer branches also have:

Field	Type	Description
transformer_hv_kv	float	High-voltage side (kV)
transformer_lv_kv	float	Low-voltage side (kV)

load — Bus Loads

Field	Type	Description
index	int	Load number
load_bus	int	Bus this load is attached to
pd	float	Active power demand (p.u.)
qd	float	Reactive power demand (p.u.)
status	int	1 = active

shunt — Reactive Compensation

Derived from DC-OPF solution to provide reactive power support for AC-OPF convergence. These are synthetic shunts — not from OSM or EIA.

Field	Type	Description
index	int	Shunt number
shunt_bus	int	Bus this shunt is attached to
gs	float	Shunt conductance (p.u.) — always 0.0
bs	float	Shunt susceptance (p.u.) — positive = capacitor, negative = reactor
status	int	1 = active

dcline — HVDC Lines

Field	Type	Description
index	int	DC line number
f_bus	int	From bus
t_bus	int	To bus
br_status	int	1 = in service
pf / pt	float	Active power at from/to end (p.u.)
qf / qt	float	Reactive power at from/to end (p.u.)
vf / vt	float	Voltage at from/to end (p.u.)
pmaxf / pmaxt	float	Max active power at from/to (p.u.)
pminf / pmint	float	Min active power at from/to (p.u.)
qmaxf / qmaxt	float	Max reactive power at from/to (p.u.)
qminf / qmint	float	Min reactive power at from/to (p.u.)
loss0	float	Constant loss coefficient
loss1	float	Linear loss coefficient
circuit_key	string	Internal circuit identifier
length_km	float	Line length (km)
model / ncost / cost		Cost model (same format as generators)

_warm_start — DC Warm-Start Metadata

Present in all released model files. Contains metadata about the DC-OPF warm-start that was applied before AC-OPF solving (voltage angles injected into buses, reactive shunts added).

Field	Type	Description
warm_start_applied	bool	Whether DC solution was injected
dc_objective	float	DC-OPF optimal cost ($/h)
dc_solved_level	int	Relaxation level at which DC converged (0 = strict)
vm_init	float	Initial voltage magnitude used (always 1.0)
n_dc_shunts	int	Number of shunts derived from DC solution
total_shunts	int	Total shunts in model

Results Files

DC Results (*_dc_results.json)

Linear DC-OPF solution. Does not solve for reactive power or voltage magnitudes.

Field	Type	Description
formulation	string	"dc"
termination_status	string	"LOCALLY_SOLVED" or "LOCALLY_INFEASIBLE"
objective	float	Total generation cost ($/h)
solve_time	float	Solver time (seconds)
relaxation_level	int	0 = strict, 1–5 = progressively relaxed
relaxation_label	string	Short label (e.g., "L0", "AC1")
relaxation_name	string	Human-readable relaxation level (e.g., "Strict")
total_gen_mw	float	Total generation (MW)
total_load_mw	float	Total load (MW)
n_buses / n_branches / n_gens / n_loads	int	Element counts
n_shunts	int	Number of shunts in model
n_decommitted	int	Generators decommitted by unit commitment
solution	dict	Per-element solutions (see below)

solution.bus: va (voltage angle, rad), vm (always 1.0 for DC)

solution.gen: pg (active power, p.u.), pg_cost (generation cost, $/h)

solution.branch: pf (from-end active flow, p.u.), pt (to-end active flow, p.u.)

solution.dcline: pf, pt, p_dc_cost

AC Results (*_ac_results.json)

Full nonlinear AC-OPF solution. Same top-level fields as DC results, plus:

Field	Type	Description
n_interfaces	int	Number of inter-BA interface constraints

solution.bus: va, vm (solved voltage magnitude)

solution.gen: pg, qg (reactive power, p.u.), pg_cost

solution.branch: pf, pt, qf, qt

solution.dcline: pf, pt, qf, qt, p_dc_cost

Interface Constraints

Models spanning multiple Balancing Authorities include an interface section with inter-BA transfer limits. Only present in multi-BA states/regions (31 of 54 datasets).

interface — Inter-BA Transfer Limits

Keyed by string ID. Each entry describes one directional interface between two BAs.

Field	Type	Description
name	string	Interface name (e.g., "PNM_to_SWPP")
from_ba	string	Source BA code
to_ba	string	Destination BA code
branch_ids	list	Branch IDs forming this interface
n_lines	int	Number of EHV lines in interface
n_all_lines	int	Total cross-BA lines (including lower voltage)
limit	float	Transfer limit (p.u.)
limit_factor	float	Fraction of total capacity used as limit
limit_method	string	"known" (NERC/WECC paths) or "heuristic"
total_rate_a	float	Sum of branch ratings (p.u.)

Data Sources

• Topology: OpenStreetMap power infrastructure (lines, substations, generators)
• Generator data: U.S. EIA-860 (capacity, fuel type, location), U.S. EIA-923 (heat rates)
• Demand: U.S. EIA-930 (hourly BA-level demand)
• Gas prices: Henry Hub spot price via EIA API
• Load allocation: U.S. Census tract-level population as spatial proxy
• BA boundaries: HIFLD Electric Planning Areas (ArcGIS FeatureServer)

Per-Unit Convention

All power quantities use system base baseMVA = 100 MVA:

• Powers (pg, pd, pmax, rate_a, etc.): multiply by 100 to get MW or MVA
• Impedances (br_r, br_x): in per-unit on system base
• Voltages (vm, vmax, vmin): in per-unit on bus base_kv
• Angles (va, angmin, angmax): in radians
• Cost coefficients: scaled for per-unit pg (i.e., cost[1] is $/h per unit of pg in p.u.)

Relaxation Levels

Models that don't converge at strict limits are progressively relaxed. For DC-OPF the solver tries L0 → L1 → … → L5. For AC-OPF the solver tries L0 → AC1 → L1 → … → L5; if AC1 alone doesn't solve it, its V/Q relaxation is kept as a base layer for L1–L5.

The OPF solver with relaxation support is available in the GridSFM repository. These models can be used directly as input.

Level	Label	Description
0	L0 — Strict	Model as-is from pipeline
1	L1 — Widen angles	Branch angles widened to ±60°
2	L2 — Thermal headroom	Branch ratings ×1.2, angles ±60°
3	L3 — Aggressive	Branch ratings ×1.5, angles ±90°, pmin ×0.5
4	L4 — Load shedding	Cap load at 70%, ratings ×1.5, angles ±90°, pmin = 0
5	L5 — Full relaxation	Remove thermal limits, angles ±90°, V [0.85, 1.15], Q ×2.0, load cap 70%, pmin = 0
AC1	AC1 — Voltage + Q	Voltage [0.90, 1.10], Q limits ×1.5 (AC-OPF only)

The relaxation_level and relaxation_label fields in results files indicate which level was needed.