Summary

In the design and operation of positive-energy buildings a pragmatic target is maximization of the actual net energy produced (NEP) by intelligently shaping demand to perform generation-consumption matching. To achieve this, informed decisions in (almost) real-time are required to operate building subsystems and to account for unpredictable user-behavior and changing weather conditions. These decisions have direct consequences to occupant thermal comfort, energy efficiency and, ultimately, to the NEP. The complex interplay between the many parameters precludes empiricism or rule-based decisions and necessitates the development of generic decision tools.

With the belief that maximization of the NEP for Positive-Energy Buildings is attained thru Better ControL dEcisions (PEBBLE), a control and optimization ICT methodology that combines model-based predictive control and cognitive-based adaptive optimization is proposed. There are three essential ingredients to the PEBBLE system: first, thermal simulation models, that are accurate representations of the building and its subsystems; second, sensors, actuators, and user interfaces to facilitate communication between the physical and simulation layers; and third, generic control and optimization tools that use the sensor inputs and the thermal models to take intelligent decisions. Building occupants have a dual sensor-actuator role in the PEBBLE framework: through user-interfaces humans act as sensors communicating their thermal comfort preferences to the PEBBLE system, and in return the PEBBLE system returns information with the goal of enhancing energy-awareness of the users. The generality of the proposed methodology affords a universality that transcends regional, behavioral, environmental or other variations. For this reason, the PEBBLE system will be demonstrated and evaluated in three buildings possessing a variety of design and performance characteristics, located at different places across Europe.

Project PEBBLE is not just about improved energy-efficiency or generation-consumption matching, it is about utilizing harmoniously, and most effectively all installed systems in a building, taking into account human factors, and adapting the decisions in (almost) real-time as and when uncertainties occur.