⚡ Quantum-AI Digital Twin — Indian Smart Grid Optimization
Hybrid Quantum-AI system for optimising EV charging in Indian smart cities. Combines AI demand forecasting + QUBO/QAOA quantum optimisation + digital twin simulation.
📊 Results
| Method | Peak Load (kW) | Peak ↓ | Avg CO₂ (g/kWh) | CO₂ ↓ | EV Cost (₹) | Renewable |
|---|---|---|---|---|---|---|
| 🔴 Baseline | 532.0 | — | 696.0 | — | ₹7451.80 | 15.5% |
| 🔵 Classical | 441.3 | 17.0% | 746.0 | -7.2% | ₹7517.10 | 9.3% |
| 🟢 Hybrid Q-AI | 441.3 | 17.0% | 746.0 | -7.2% | ₹7517.10 | 9.3% |
AI Forecast (full year): LSTM MAE=3.7536 GW · TFT MAE=3.4121 GW · Ensemble R²=0.9297
🗃️ Indian Datasets
| Dataset | Kaggle | Records |
|---|---|---|
| Indian Power Consumption | anikannal | 8,760 |
| Solar Power Generation | anikannal | 17,520 |
| EV Charging Stations | piyushagni5 | 7,300 |
| Weather India | sudalairajkumar | 8,760 |
🏗️ Pipeline
Indian Datasets → Feature Engineering (26 features) →
AI Forecasting (BiLSTM+Attention 1.25M params | TFT 237K params | Ensemble) →
Digital Twin (Local Substation · ToU ₹/kWh · Carbon Tracking) →
QUBO → QAOA (Qiskit) → Hybrid Solver → Optimal EV Schedule
📄 Paper Title
"Hybrid Quantum-AI Digital Twin for Indian Smart Grid EV Charging Optimization using Renewable-Aware Scheduling" Target: IEEE Trans. Smart Grid | Applied Energy | Energies (MDPI)
Generated: 2026-03-13 15:53 UTC
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support