Real long-term performance evaluation of an improved office building operation involving a Data-driven model predictive control

• Data-driven Model Predictive Control (DMPC) uses an adaptive grey-box model. • Energy Conservation Measures (ECMs), including DMPC, are tested for one year. • The ECMs are evaluated with the Measurement and Verification (M&V) method. • The DMPC, evaluated separately, jointly controls heating/cooling and shading. • The DMPC saves significant thermal energy and enables demand side flexibility. Data-driven Model Predictive Control (DMPC) strategies, coupled with holistically optimized HVAC system control, represent a promising approach to achieve climate targets through significant reductions in building energy consumption and associated emissions. To validate this potential in a real-world environment, a comprehensive optimization study was conducted on an office building serving as a living laboratory. Through systematic analysis of historical operational data, multiple Energy Conservation Measures (ECMs) were identified and implemented. The cornerstone of these improvements was the development and deployment of a centralized adaptive DMPC system, which was operated and evaluated over a full year. The implemented system employs reduced-order state-space (resistance–capacitance) models for each building zone, complemented by simplified representations of storage tanks and heat pump dynamics. By integrating weather forecasting with a 24-hour prediction horizon into the control strategy, the DMPC simultaneously manages heating or cooling loads and shading blinds to optimize solar gain utilization. This integrated approach to controlling the primary heat fluxes enables advanced load shifting and peak shaving capabilities through automated management. The implementation yielded substantial technological advantages in HVAC operation: the DMPC algorithm’s effective peak shaving significantly reduced maximum load requirements, enabling the complete elimination of the gas boiler. This reduction in peak demands made it possible for the heat pump to operate efficiently, maintaining full operational capabilities with just a small-sized ground source heat exchanger, instead of the conventional air source configuration. This demonstrates how the DMPC system successfully leveraged the building’s demand-side flexibility potential, yielding improved operational economics and extended equipment service life. When compared to a baseline year of operation utilizing rule-based control for heating/cooling and partial automation of shading, these Energy Conservation Measures (ECMs), including the DMPC implementation, demonstrated remarkable energy savings. Following adjustment through the Measurement and Verification (M&V) framework, the analysis revealed that the DMPC alone achieved reductions of 41% in overall thermal energy consumption in the zones, comprising 17% in heating and 75% in cooling. The DMPC contributed 30% of the total thermal energy savings achieved across all implemented ECMs. Collectively, the ECMs resulted in a 54% reduction in total thermal energy consumption (38% heating, 80% cooling), while final energy consumption for heating and cooling decreased by 85%. These findings underscore the efficacy of DMPC in practical applications and its capacity to significantly advance building energy efficiency and sustainability objectives. The study demonstrates that integrating DMPC within holistic improvement strategies represents a viable approach to achieve both environmental targets and operational enhancements in the building sector.