In light of the growing strain on the energy grid and the increased awareness of the significant role buildings play within the energy ecosystem, the need for building operational strategies which minimize energy consumption has never been greater. One of the major hurdles impeding this realization primarily lies not in the lack of decision strategies, but in their inherent lack of adaptability. With most operational strategies partly dictated by a dynamic trio of social, economic and environmental factors which include occupant preference, energy price and weather conditions, it is important to realize and capitalize on this dynamism to open up new avenues for energy savings. This paper extends this idea by developing a dynamic optimization mechanism for Net-zero building clusters. A bi-level operation framework is presented to study the energy tradeoffs resulting from the adaptive measures adopted in response to hourly variations in energy price, energy consumption and indoor occupant comfort preferences. The experimental results verify the need for adaptive decision frameworks and demonstrate, through Pareto analysis, that the approach is capable of exploiting the energy saving opportunities made available through fluctuations in energy price and occupant comfort preferences.
This work leverages the current state of the art in reinforcement learning for continuous control, the Deep Deterministic Policy Gradient (DDPG) algorithm, towards the optimal 24-hour dispatch of shared energy assets within building clusters. The modeled DDPG agent interacts with a battery environment, designed to emulate a shared battery system. The aim here is to not only learn an efficient charged/discharged policy, but to also address the continuous domain question of how much energy should be charged or discharged. Experimentally, we examine the impact of the learned dispatch strategy towards minimizing demand peaks within the building cluster. Our results show that across the variety of building cluster combinations studied, the algorithm is able to learn and exploit energy arbitrage, tailoring it into battery dispatch strategies for peak demand shifting.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.