TY - GEN
T1 - Data-Driven Integration of Thermal Comfort and Demand-Side Management Using PV Measurements in Qatar
AU - Al Qahtani, Muneera
AU - Boumaiza, Ameni
N1 - Publisher Copyright:
© 2026 IEEE.
PY - 2026
Y1 - 2026
N2 - Cooling demand accounts for the largest share of electricity use in hot arid climates and places a heavy burden on grid stability. To address this challenge, we propose a data-driven framework that brings together thermal comfort modeling and Demand-Side Management (DSM) using real photovoltaic (PV) data from Education City, Qatar. At the heart of the proposed framework lies a stochastic model-predictive control (MPC) strategy, engineered to facilitate the harmonious operation of diverse building energy systems. This encompasses the synchronized management of HVAC (heating, ventilation, and air conditioning) systems, efficient utilization of photovoltaic (PV) energy, and precise scheduling of battery storage. Paramount to this strategy is the continuous preservation of occupant comfort, rigorously evaluated against the Predicted Mean Vote (PMV) and Predicted Percentage of Dissatisfied (PPD) metrics. Short-term PV forecasts are generated using a Gradient-Boosted Regression Tree (GBRT) model trained on high-resolution solar and PV datasets. Results show that the integrated PV-battery MPC strategy lowers cooling energy consumption by 18.7% and peak grid imports by 20%, all while keeping comfort within ASHRAE standards for more than 98% of occupied hours. This work demonstrates a practical pathway for occupant-centric DSM in Qatar and similar regions, advancing both energy efficiency and grid resilience without sacrificing comfort.
AB - Cooling demand accounts for the largest share of electricity use in hot arid climates and places a heavy burden on grid stability. To address this challenge, we propose a data-driven framework that brings together thermal comfort modeling and Demand-Side Management (DSM) using real photovoltaic (PV) data from Education City, Qatar. At the heart of the proposed framework lies a stochastic model-predictive control (MPC) strategy, engineered to facilitate the harmonious operation of diverse building energy systems. This encompasses the synchronized management of HVAC (heating, ventilation, and air conditioning) systems, efficient utilization of photovoltaic (PV) energy, and precise scheduling of battery storage. Paramount to this strategy is the continuous preservation of occupant comfort, rigorously evaluated against the Predicted Mean Vote (PMV) and Predicted Percentage of Dissatisfied (PPD) metrics. Short-term PV forecasts are generated using a Gradient-Boosted Regression Tree (GBRT) model trained on high-resolution solar and PV datasets. Results show that the integrated PV-battery MPC strategy lowers cooling energy consumption by 18.7% and peak grid imports by 20%, all while keeping comfort within ASHRAE standards for more than 98% of occupied hours. This work demonstrates a practical pathway for occupant-centric DSM in Qatar and similar regions, advancing both energy efficiency and grid resilience without sacrificing comfort.
KW - Demand-Side Management
KW - Photovoltaic Data
KW - Qatar
KW - Smart Thermostat
KW - Thermal Comfort
UR - https://www.scopus.com/pages/publications/105053342871
U2 - 10.1109/PESGM58988.2026.11693239
DO - 10.1109/PESGM58988.2026.11693239
M3 - Conference contribution
AN - SCOPUS:105053342871
T3 - IEEE Power and Energy Society General Meeting
BT - 2026 IEEE Power and Energy Society General Meeting, PESGM 2026
PB - IEEE Computer Society
T2 - 2026 IEEE Power and Energy Society General Meeting, PESGM 2026
Y2 - 19 July 2026 through 23 July 2026
ER -