AI-Driven Digital Twins for Indoor Air Quality and Energy Optimization in Smart Buildings: A Simulated Control Study

This article presents a simulated empirical assessment in Civil Engineering, Building Science, and Artificial Intelligence. The simulation generated 40,320 zone-time observations and included weekday occupancy variability, meeting-room surges, outdoor pollution episodes, and three modeled sensor-fault events. A simulated five-zone office building was modeled over a 12-week cooling-season period at 15-minute intervals. Inputs included occupancy, outdoor temperature, relative humidity, outdoor PM2.5, internal heat gains, ventilation flow, and HVAC power. A digital twin maintained zone-level state estimates and used a short-horizon predictive controller to adjust outside-air delivery and cooling setpoints. The study compared fixed schedules, single-sensor CO2 demand control, and multi- variable digital-twin control. Digital-twin control reduced simulated HVAC electricity by 16.7% relative to fixed scheduling while decreasing occupied-hours above 1,000 ppm CO2 from 14.9% to 4.8%. It also responded more effectively to outdoor PM2.5 episodes by temporarily reducing untreated outside-air intake and increasing recirculation filtration. Thermal discomfort hours remained below 5% in all strategies. The numerical values are synthetic and are used solely to demonstrate a plausible empirical design, comparative analysis, and journal-style reporting workflow. No real participants, operational institutions, commercial facilities, or field sites were studied. The analysis emphasizes current research trends, multi-metric evaluation, practical trade-offs, and the importance of validation before deployment.