QuantumGridOS

QuantumGridOS is a Python library for connecting power systems data to quantum computers.

It enables solving power system optimization problems using quantum algorithms (QAOA, VQE) with minimal latency TCP/IP data exchange.

🚀 Features

• Real-time TCP/IP streaming for power systems data exchange
• Quantum algorithm support: QAOA, VQE, Grover's
• Power system optimizations: Unit commitment, MaxCut, OPF, State estimation
• Low-latency architecture with async I/O and buffer management
• Hardware agnostic: Works with IBM Quantum, IonQ, Rigetti, simulators
• Extensive examples and documentation

📦 Installation

pip install quantumgridos

For development:

git clone https://github.com/saralsystems/quantumgridos.git
cd quantumgridos
pip install -e .[dev]

🎯 Quick Start

Basic MaxCut for Power Network Partitioning

import quantumgridos as qgo

# Initialize quantum-power interface
interface = qgo.QuantumPowerInterface(
    quantum_backend='qiskit_aer',
    tcp_host='localhost',
    tcp_port=5000
)

# Define power network
network = qgo.PowerNetwork.from_ieee_case(14)  # IEEE 14-bus

# Create MaxCut optimizer for network partitioning
optimizer = qgo.MaxCutOptimizer(
    network=network,
    algorithm='qaoa',
    layers=3
)

# Start real-time processing
async def process_stream():
    async for data in interface.tcp_stream():
        # Solve partitioning problem
        result = await optimizer.solve_async(data)
        
        # Send result back to power system
        await interface.send_result(result)

# Run
import asyncio
asyncio.run(process_stream())

Unit Commitment Example

import quantumgridos as qgo

# Configure unit commitment problem
uc_problem = qgo.UnitCommitment(
    generators=[
        {'name': 'G1', 'pmin': 50, 'pmax': 200, 'cost': 1000},
        {'name': 'G2', 'pmin': 20, 'pmax': 100, 'cost': 1500}
    ],
    demand_forecast=[150, 180, 200, 170],
    time_periods=4
)

# Setup quantum solver
solver = qgo.QuantumSolver(
    problem=uc_problem,
    backend='ibmq_qasm_simulator',
    algorithm='vqe',
    optimizer='cobyla'
)

# Solve with TCP streaming
with qgo.TCPInterface(port=5000) as tcp:
    for demand_update in tcp.stream():
        uc_problem.update_demand(demand_update)
        solution = solver.solve()
        tcp.send(solution.to_scada_format())
        solution = solver.solve()
        tcp.send(solution.to_scada_format())

Low-Inertia Counterfactual Search

import quantumgridos as qgo

# Built-in Braket-scale demo case inspired by the GB 9 Aug 2019 disturbance.
study = qgo.create_low_inertia_study()

# QuantumGridOS builds a QUBO candidate-search layer, then validates every
# candidate with reduced frequency-security physics.
result = qgo.solve_low_inertia_counterfactual(
    study,
    solvers=("exact", "annealing", "qaoa"),
    top_k=24,
    qaoa_layers=1,
)

print(result.best.portfolio.selected_labels)
print(result.best.metrics["nadir_hz"])
print(result.best.metrics["max_abs_rocof_hz_s"])

Use this API for low-inertia studies where quantum or quantum-inspired solvers search over approved binary grid/DER settings such as synchronous commitment, BESS fast-frequency response, grid-forming mode, DER ride-through package, and pre-event export reduction. Classical physics validation remains authoritative.

For publication screening, enable the native two-area dynamic-label surrogate:

result = qgo.solve_low_inertia_counterfactual(
    study,
    solvers=("exact", "annealing"),
    dynamic_labels=True,
)

label = qgo.label_low_inertia_dynamics(study, bits=result.best.portfolio.bitstring)
print(label.metrics["max_relative_angle_rad"])
print(label.metrics["min_electromechanical_damping_ratio"])

The dynamic labeler adds inter-area angle, area-frequency split, and small-signal damping checks. It is a reproducible reduced electromechanical screen, not a replacement for RMS/EMT validation.

Public handoff links and CUDA execution requirements are available directly from the package:

import quantumgridos as qgo

catalog = qgo.get_low_inertia_public_data_catalog()
print(catalog["s3_public_handoff_bundle"]["s3_uri"])
print(catalog["public_sources"]["neso_august_2019_frequency"])

The July 2026 handoff bundle is stored at:

• S3 URI: s3://qgo-low-inertia-public-data-20260706-654777652612/public/low_inertia_quantum_public_handoff_20260706.zip
• SHA256: 534e4c6b01845f3ac9e48d4fb3f4b0104db025b2f4f6db833eef26ed65bc3fc2
• Approximate size: 91 MB

The object currently requires a presigned HTTPS URL. Do not commit expiring S3 signed URLs to source control; share them out-of-band or generate a fresh one from the owning AWS account. CUDA/cuOpt comparisons should use the same package-variable schema as the CPU and Qiskit solvers and report end-to-end wall-clock time.

Quantum Power Flow

import quantumgridos as qgo

# Load network from CSV
net = qgo.create_network("examples/test_case_4bus", type='csv')

# Run Quantum Newton-Raphson (HHL Fast Mode)
success, x, history, circuit = qgo.run_quantum_nr(net, method='hhl_fast')

if success:
    print("Converged! Voltage Angles:", net.buses['v_ang'].values)
    # Optional: Visualize the circuit
    # from quantumgridos.utils.visualizer import draw_circuit
    # draw_circuit(circuit)

🏗️ Architecture

QuantumGridOS/
├── Core Modules
│   ├── quantum_interface.py     # Quantum backend abstraction
│   ├── tcp_handler.py          # High-performance TCP/IP
│   ├── data_encoder.py         # Power data → Qubits
│   └── time_sync.py            # Clock synchronization
├── Algorithms
│   ├── qaoa.py                 # QAOA implementation
│   ├── vqe.py                  # VQE implementation
│   └── grover.py               # Grover's algorithm
├── Power Systems
│   ├── network.py              # Power network modeling
│   ├── optimizations/
│   │   ├── unit_commitment.py
│   │   ├── opf.py             # Optimal Power Flow
│   │   ├── state_estimation.py
│   │   └── maxcut.py
│   └── converters.py           # IEEE/MATPOWER formats
└── Utils
    ├── benchmarks.py
    └── visualization.py

📊 Benchmarks

🔌 TCP/IP Protocol

QuantumGridOS uses optimized binary protocol for minimal latency:

# Message format
{
    'timestamp': int64,          # Unix timestamp in microseconds
    'msg_type': uint8,          # 0: data, 1: control, 2: result
    'data': {
        'bus_voltages': float32[],
        'line_flows': float32[],
        'generator_status': bool[]
    }
}

📚 Documentation

Full documentation available at saralsystems.github.io/quantumgridos

• Getting Started Guide
• Quick Start
• API Reference
• User Guide
• Tutorials
• Quantum Power Flow Guide

🧪 Testing

# Run all tests
pytest tests/

# Run with coverage
pytest --cov=quantumgridos tests/

# Run benchmarks
python -m quantumgridos.benchmark

🤝 Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

All contributors must sign a Contributor License Agreement (CLA) before their contributions can be merged. See CLA.md for individual contributors and CLA-CORPORATE.md for corporate contributors.

📄 License

Apache License 2.0 - see LICENSE file.

📖 Citation

If you use QuantumGridOS in research, please cite:

@software{quantumgridos,
  title = {QuantumGridOS: Real-time Quantum-Power Systems Interface},
  author = {Saral Systems},
  year = {2025},
  url = {https://github.com/saralsystems/quantumgridos},
  license = {Apache-2.0}
}

🙏 Acknowledgments

Based on research from NREL ARIES and quantum-in-loop (QIL) architecture.