cs.LGNov 18, 2026

From Data to Valves: A Self-Learning, Production-Grade Edge Stack for Sustainable HVAC Control

Authors: Abbas Raza Ali, Muhammad Hamza Rehman, Xu Li, Muhammad Ajmal Siddiqui

Organizations: Citigrounp Inc., London, United Kingdom · National University of Computing & Emerging Sciences, Islamabad, Pakistan · Alibaba Cloud, Amsterdam, Netherlands · Ernst & Young LLP, London, United Kingdom

Abstract

Learned HVAC controllers rarely leave the simulator, and those that do are typically trained offline, frozen at deployment and retrained from the cloud. We present a production stack in which three learning components run on one NVIDIA Jetson Orin gateway per building, adapt continuously from the building's own data, and verify their own effect. (a) A compact transformer-encoder forecaster, one per equipment group with learned per-unit embeddings, predicts every air-side unit's next-cycle temperature and learns prequentially, one gradient step per five-minute cycle, behind five independent gates that bound the influence of any corrupt reading. (b) A distributional soft actor-critic agent, one shared policy across unit types, converts the measured state into a chilled-water valve position under a structural safety envelope that no learned component can soften; its second generation learns nightly on the device from recorded transitions alone, with a behaviour anchor and critic calibration terms, is scored by four model-free metrics and promoted by atomic hot-swap. (c) A change-point regression baseline held to ASHRAE Guideline 14 supplies the counterfactual against which savings are certified. On eighteen months of data from a hotel the forecaster converges from random weights within a week and reaches 0.37, 0.39 and 0.11 degree C MAE for air-handling, fresh-air and fan-coil units with no retraining event; on a different hotel the baseline reproduces a 46-day held-out period to within 0.07%. The RL agent has completed its in-simulator campaign and runs in shadow on 104 units; we report measured latency and memory of every service on a shared 15 GB device, the systems lessons of the campaign, and precisely what the evidence does and does not yet establish.

Figures & tables

Explore similar work

CardsList
  1. When does deep reinforcement learning pay off in HVAC supervisory control? Four small Modelica case studies trained directly on FMUs

    Nov 17, 2026Bingnan Zhao, Harshala Gammulle, Tharindu FernandoHeating, Ventilation, and Air Conditioning ControlBuilding Energy Management

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

    Oct 14, 2026Panagiotis Kasnesis, Lazaros Toumanidis, Christos Chatzigeorgiou +2Heating, Ventilation, and Air Conditioning ControlBuilding Energy Management

  3. Generalizing HVAC Control With Domain Randomized Reinforcement Learning

    Sep 5, 2026Pablo Boitel, Kun ZhangHeating, Ventilation, and Air Conditioning ControlRL Control