arXivDaily arXiv每日学术速递 周一至周五更新
看起来今天可以划水了
arXiv 2103.07919cs.AIcs.SYeess.SY
收藏

基于深度强化学习的人机交互建筑控制仿真研究

Simulation Studies on Deep Reinforcement Learning for Building Control with Human Interaction

KAIST(韩国科学技术院);ETH Zürich(苏黎世联邦理工学院);Purdue University(普渡大学)

Donghwan Lee, Niao He, Seungjae Lee, Panagiota Karava, Jianghai Hu

摘要:

建筑业是世界上消耗能源最多的行业,因此对建筑物的能耗和舒适度管理有着相当大的研究兴趣。受强化学习(RL)最新进展的启发,本文旨在评估RL在具有居住者交互的建筑气候控制问题中的潜力。我们采用了最近的RL方法,称为DDPG(深确定性策略梯度),并通过仿真研究评估其处理以下问题的能力:(a)由于传感器限制而导致的部分状态可观测性;(b)具有高维状态空间的复杂随机系统,这些状态空间是连续和离散的;(c)由于周围天气条件、居住者的行为和舒适感而引起的不确定性。特别是,部分可观性和不确定性,由于乘员的相互作用显着复杂的控制问题。通过仿真研究,DDPG学习的策略表现出合理的性能和计算的易处理性。

英文摘要:

The building sector consumes the largest energy in the world, and there have been considerable research interests in energy consumption and comfort management of buildings. Inspired by recent advances in reinforcement learning (RL), this paper aims at assessing the potential of RL in building climate control problems with occupant interaction. We apply a recent RL approach, called DDPG (deep deterministic policy gradient), for the continuous building control tasks and assess its performance with simulation studies in terms of its ability to handle (a) the partial state observability due to sensor limitations; (b) complex stochastic system with high-dimensional state-spaces, which are jointly continuous and discrete; (c) uncertainties due to ambient weather conditions, occupant's behavior, and comfort feelings. Especially, the partial observability and uncertainty due to the occupant interaction significantly complicate the control problem. Through simulation studies, the policy learned by DDPG demonstrates reasonable performance and computational tractability.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑