Low Emission Building Control with Zero-Shot Reinforcement Learning

TL;DR

This paper introduces PEARL, a zero-shot reinforcement learning approach that reduces building emissions by up to 31% without requiring building-specific data or simulators, improving energy efficiency and maintaining comfort.

Contribution

The paper presents PEARL, a novel zero-shot RL method combining system identification and model-based RL for emission reduction without prior building data.

Findings

PEARL outperforms rule-based controllers and RL baselines in simulations.

PEARL reduces emissions by up to 31%.

A short exploration phase suffices to build effective models.

Abstract

Heating and cooling systems in buildings account for 31\% of global energy use, much of which are regulated by Rule Based Controllers (RBCs) that neither maximise energy efficiency nor minimise emissions by interacting optimally with the grid. Control via Reinforcement Learning (RL) has been shown to significantly improve building energy efficiency, but existing solutions require access to building-specific simulators or data that cannot be expected for every building in the world. In response, we show it is possible to obtain emission-reducing policies without such knowledge a priori--a paradigm we call zero-shot building control. We combine ideas from system identification and model-based RL to create PEARL (Probabilistic Emission-Abating Reinforcement Learning) and show that a short period of active exploration is all that is required to build a performant model. In experiments across three varied building energy simulations, we show PEARL outperforms an existing RBC once, and popular RL baselines in all cases, reducing building emissions by as much as 31\% whilst maintaining thermal comfort. Our source code is available online via https://enjeeneer.io/projects/pearl .