A Data-driven Augmented TABS Control Strategy in a Smart Building for the South-facing Offices

Abstract

The thermally activated building system (TABS) is becoming a popular heating option in buildings in the US. A typical TABS heating control strategy uses an embedded slab temperature sensor or room air temperature sensor as the control reference. TABS control may be disturbed by heat gain in the room space, for example, the intense solar radiation, due to the thermal inertia of the thermal mass slab. Overheating is often observed in the south-facing offices in winter. The fundamental problem is that the TABS heating control design is only at the system level, not considering the whole building and the built environment. This paper proposes a novel data-driven control strategy to augment the existing TABS heating control strategy considering future weather conditions. The TABS and solar energy are modeled in a monolithic energy model for the first time through a data-driven method. Furthermore, this data-driven control strategy has been successfully evaluated in a smart building in Cambridge, Massachusetts.