arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.19790cs.MAcs.AI
收藏

每个区域一个智能体:边缘硬件上多区域HVAC控制的部分可观测去中心化主动推理

One Agent per Zone: Decentralised Active Inference under Partial Observability for Multi-Zone HVAC Control on Edge Hardware

Waldiez P.C.(Waldiez公司);ThinGenious P.C.(ThinGenious公司);PRAESC AI(PRAESC人工智能)

Panagiotis Kasnesis, Lazaros Toumanidis, Christos Chatzigeorgiou, Amalia Contiero Syropoulou, Miguel De Prado

AI总结:

本研究提出一种为多区域HVAC控制设计的去中心化主动推理智能体集群,在边缘硬件上实现了更低的建筑峰值功率与错开的楼层峰值,性能优于基准方法。

摘要:

多区域HVAC控制是一个共享资源问题:楼层上的各个区域共享一台风机和冷却盘管,因此每个区域的需求会影响相邻区域以及建筑的峰值功率。为每个区域配备一个控制器可实现良好的扩展性并隔离故障,但此时每个控制器仅能观测自身区域,无法观测共享空气回路。本研究提出一种去中心化控制器,为每个区域分配一个独立的主动推理(active inference)智能体,该智能体通过最小化预期自由能进行规划,并在线学习自身区域的动态特性。智能体之间无需交换任何消息,因此无需依赖其他智能体或协调器;每个智能体通过潜在干扰项和从其他楼层计量功率推导得到的拥塞信号来解释未观测到的耦合关系。在Sinergym办公建筑的EnergyPlus仿真环境中,对15个智能体进行评估,对比对象包括3个基于规则的控制器和3个MADDPG集群(其中一个集群配备相同的楼层计量设备)。在训练期间未见过的5个含噪声天气年份中,该集群实现了最低的建筑平均峰值功率(88.3 kW,比基准低3%至17%),且比任何基准都更能错开楼层峰值,其重合因子(建筑峰值与楼层峰值总和的比值)为0.929,而基准的重合因子为0.936至0.994。在舒适度相近的情况下(0.912对比0.916),与反应式规则相比,该方法降低了每月峰值总和(即计费需求费用的依据),每年减少33 kW(4%),代价是能耗增加约10%。在每个测试年份中,其年度成本比主动式规则(舒适度较低)低9%,比配备相同楼层计量设备的MADDPG集群低1%。整个集群可在单个低成本Raspberry Pi 5上运行,通过MQTT完成观测到动作的闭环时间在p95时为442 ms,远低于15分钟的控制步长。

英文摘要:

Multi-zone HVAC control is a shared-resource problem: zones on a floor share one fan and cooling coil, so each zone's demand affects its neighbours and the building's peak power. One controller per zone scales well and isolates faults, but each controller then observes only its own zone, not the shared air loop. In this work, we present a decentralised controller that assigns each zone an independent active-inference agent, which plans by minimising expected free energy and learns its zone dynamics online. The agents exchange no messages, so none depends on peers or a coordinator; each accounts for the unobserved coupling through a latent disturbance and a congestion signal derived from the metered power of the other floors. Fifteen agents are evaluated in EnergyPlus on a Sinergym office building against three rule-based controllers and three MADDPG fleets, one given the same floor meters. On five noisy weather years unseen during training, the fleet attains the lowest mean building peak (88.3 kW, 3 to 17 % below the baselines) and staggers the floor peaks better than any baseline, with a coincidence factor (building peak over the sum of the floor peaks) of 0.929 versus 0.936 to 0.994. Compared with a reactive rule at similar comfort (0.912 versus 0.916), it lowers the sum of monthly peaks, on which demand charges are billed, by 33 kW (4 %) in every year, at the cost of about 10 % more energy. It costs 9% less per year than the proactive rule (at lower comfort) and 1% less than the MADDPG fleet given the same floor meters, in every test year. The whole fleet runs on a single low-cost Raspberry Pi 5, closing the observation-to-action loop over MQTT in 442 ms at p95, far below the 15-minute control step.

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。

↑