每区一智能体：边缘硬件上多区HVAC去中心化主动推理控制
原标题：One Agent per Zone: Decentralised Active Inference under Partial Observability for Multi-Zone HVAC Control on Edge Hardware
AI 摘要
该论文提出一种去中心化主动推理（AIF）控制器，为多区域HVAC系统中每个区域分配一个独立智能体，各智能体仅观测本区域、不互相通信，通过最小化期望自由能进行规划并在线学习区域动态，同时利用潜在扰动和来自其他楼层计量功率的拥塞信号应对未观测耦合。在EnergyPlus/Sinergym办公楼环境中，15个智能体与三种规则控制器和三种MADDPG方案对比，在五个未见天气年份上取得最低建筑峰值（比基线低3%至17%），重合因子0.929优于基线0.936至0.994，月度峰值之和降低约4%，代价是能耗增加约10%。整个智能体集群运行在单台低成本Raspberry Pi 5上，通过MQTT闭环，p95延迟远低于15分钟控制步长。
正文节选
One Agent per Zone: Decentralised Active Inference under Partial Observability for Multi-Zone HVAC Control on Edge Hardware Abstract 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 t