Safety-Constrained Hybrid Quantum-Classical Optimization for Microgrid Energy Management
SafeGrid-Q is an experimental project investigating optimization methods for energy scheduling in small microgrids. The project explores classical optimization and QAOA (Quantum Approximate Optimization Algorithm) for scheduling electrical loads under power constraints.
The long-term goal is to investigate scheduling scenarios involving EV charging, solar generation, and battery storage while preserving deterministic safety constraints.
- QAOA-based binary load scheduling
- QUBO and Ising formulations
- Comparison with exact enumeration and mixed-integer linear programming (MILP)
- Multi-period energy scheduling
- Solar-aware load scheduling
- Battery state-of-charge and charging/discharging constraints
- Feasibility, solution quality, circuit complexity, and noise-sensitivity experiments
SafeGrid_Q.ipynb— notebook containing the current experiments, explanations, and results.
- Python
- Qiskit
- NumPy and SciPy
- Classical optimization
- QUBO, Ising Hamiltonians, and QAOA
The experiments are research prototypes and simplified simulations. The current QAOA results do not establish quantum advantage or demonstrate deployment in a physical microgrid. More realistic constraints, larger benchmarks, and comparisons against strong classical baselines are needed.
- Compare different QAOA depths and repeated optimization runs
- Extend scheduling to multiple time slots
- Improve EV charging and battery models
- Investigate uncertain solar generation and electricity prices
- Evaluate performance against classical optimization methods
Work in progress.