QGreenTwin+: A Federated 3D Digital Twin and AI Platform for Campus-Wide Renewable Forecasting, PV Tracking, and Climate Intelligence in Qatar Foundation

Qatar Foundation’s campus can be a regional model for sustainable infrastructure and digital innovation. Building on the Q-Green energy-trade pilot, QGreenTwin+ becomes a federated 3D digital-twin for campus-wide forecasting, optimization, and resilience. It fuses live data from PV arrays, building meters, weather, and environmental sensors, using QF’s Power BI dashboard as the primary interface. Beyond prosumer trading, it targets system performance: a 3D PV tracking module simulates solar angles, shading, and weather to dynamically tune tilt/orientation, boosting yield 10–15%. An AI predictive-maintenance layer for HVAC fans, pumps, and motors flags anomalies and guides technicians via AR, cutting unplanned downtime 30–50% and O&M costs ~20%. To protect privacy and scale, federated ML keeps sensitive building data local while models learn collaboratively, reaching >90% accuracy for electricity-demand forecasts and 85–95% for PV generation. Decision-makers get intuitive views of real-time performance, carbon emissions, resilience KPIs, and scenario simulations for dust storms and heatwaves; field staff use a mobile AR overlay mapping live asset data onto equipment, shortening response times ~25–40%. Expected outcomes: an operational federated twin, dynamic PV optimization, high-accuracy forecasts, maintenance gains, AR diagnostics, and dashboards for carbon tracking and resilience planning, with KPIs of ≥10% CO₂ reduction and 8–12% energy savings.

Hamad Bin Khalifa University (HBKU)