[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83759-en":3,"doc-seo-83759-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},83759,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","ThermoForce A Physics-Structured Interventional World Model for Building HVAC Control","Model predictive control (MPC) for building HVAC requires a thermal model answering causal, counterfactual questions about indoor temperature, energy use, and comfort under specific control actions. While time-series foundation models (TSFMs) excel at zero-shot passive forecasting, high forecast accuracy does not guarantee correct responses to interventions. ThermoForce separates a frozen TSFM free-response prior from a learned, physics-structured forced-response operator whose control effect is monotone by construction, enabling recovery of intervention effects with limited control excitation. Embedded MPC on BOPTEST reduces thermal discomfort 33–84% while lowering energy, surpassing multiple baselines.","arXiv :2607 .03942v1 [ ee ss . SY] 4 Jul 2026  \nHighlights  \nThermoForce: A Physics-Structured Interventional World Model for Building HVAC Control  \nYifan Wang  \n• Factual forecasting accuracy does not imply valid HVAC control response.  \n• ThermoForce separates a frozen free-response prior from a forced-response operator.  \n• The forced operator is monotone in the control input by construction.  \n• It recovers intervention effects from one to three days of control excitation.  \n• MPC deployment cuts BOPTEST discomfort 33–84% while reducing energy.  \nThermoForce: A Physics-Structured Interventional World Model for Building HVAC Control  \nYifan Wang∗  \nDepartment of Mechanical Engineering, McGill University, 817 Sherbrooke Street West, Montreal, H3A 2T7, QC, Canada  \nAbstract  \nModel predictive control (MPC) of building heating, ventilation, and airconditioning (HVAC) systems depends on a thermal model that can answer a fundamentally causal question: what will the indoor temperature, energy use, and comfort be if a given control action is applied? Time-series foundation models (TSFMs) now forecast passive building thermal trajectories with remarkable zero-shot skill, and it is tempting to treat them as readymade thermal models for control. We show that this is unsafe: high factual forecasting accuracy does not imply valid response to control interventions. An observational grey-box model with the best passive accuracy predicts the effect of cooling actions with the wrong sign, and feeding control and weather covariates to a TSFM degrades rather than improves its intervention response. We introduce ThermoForce, a control-ready interventional thermal world model that keeps a TSFM frozen as a passive free-response prior and learns a compact, physics-structured forced-response operator for the causal effect of HVAC actuation. The operator is monotone in the control input by construction, is identified from one to three days of control excitation, and composes with the free response into a counterfactual-capable world model. Across paired EnergyPlus heating and cooling interventions, ThermoForce attains the lowest intervention-effect error and correct effect sign where covariate-TSFM, observational grey-box, and distillation baselines fail. Embedded in MPC on the BOPTEST benchmark, it reduces thermal discomfort by 33–84% relative to the native controller across three two-week  \n∗ Corresponding author.  \nEmail address: [yifan.wang18@mail.mcgill.ca](yifan.wang18@mail.mcgill.ca) (Yifan Wang)  \nwindows while simultaneously reducing energy, using a frozen backbone, 195 trainable parameters, and central-processing-unit-only computation. ThermoForce reframes foundation models for building control: passive prediction and forced intervention response must be structurally separated for a model to be control-ready.  \nKeywords: Building HVAC control, Model predictive control, Time-series foundation models, Physics-informed machine learning, Interventional world model, Counterfactual prediction  \n1. Introduction  \nBuildings account for a large share of global final energy use, and their heating, ventilation, and air-conditioning (HVAC) systems are the dominant controllable load. Model predictive control (MPC) is the most successful advanced strategy for operating these systems, repeatedly demonstrating energy and comfort improvements over rule-based control [1, 2] . The effectiveness of MPC hinges almost entirely on one component: a thermal model that can predict how the building will respond to candidate control actions over a planning horizon. The controller does not need to know only what the temperature will be; it needs to know what the temperature, energy, and comfort would be under each action it is considering. This is a causal, counterfactual query, and it is what separates a control-ready model from an ordinary forecaster.  \nTime-series foundation models (TSFMs) such as Chronos [3] and TimesFM [4] have recently transformed general-pur","cbCaibxDsWW68LJS","https://ap.wps.com/l/cbCaibxDsWW68LJS","pdf",695354,4,1,23,"English","en",105,"# Highlights\n# Abstract\n# Keywords\n# Introduction","[{\"question\":\"Why is factual forecasting accuracy insufficient for HVAC control?\",\"answer\":\"Because HVAC control is an intervention rather than a passive covariate shift. A model may predict passive thermal trajectories accurately yet produce incorrect or even sign-inverted effects when control actions are applied.\"},{\"question\":\"What is ThermoForce and how does it differ from using a TSFM directly?\",\"answer\":\"ThermoForce is a control-ready interventional thermal world model. It keeps a TSFM frozen as a passive free-response prior and learns a compact physics-structured forced-response operator to capture the causal effect of HVAC actuation.\"},{\"question\":\"How does ThermoForce ensure correct intervention response with the learned operator?\",\"answer\":\"The forced-response operator is constructed to be monotone in the control input, and it is identified from one to three days of control excitation. Together with the free response, it forms a counterfactual-capable world model for MPC.\"}]",1784190257,58,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"thermoforce-a-physics-structured-interventional-world-model-for-building-hvac-control","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/thermoforce-a-physics-structured-interventional-world-model-for-building-hvac-control/83759/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is factual forecasting accuracy insufficient for HVAC control?","Question",{"text":75,"@type":76},"Because HVAC control is an intervention rather than a passive covariate shift. A model may predict passive thermal trajectories accurately yet produce incorrect or even sign-inverted effects when control actions are applied.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is ThermoForce and how does it differ from using a TSFM directly?",{"text":80,"@type":76},"ThermoForce is a control-ready interventional thermal world model. It keeps a TSFM frozen as a passive free-response prior and learns a compact physics-structured forced-response operator to capture the causal effect of HVAC actuation.",{"name":82,"@type":73,"acceptedAnswer":83},"How does ThermoForce ensure correct intervention response with the learned operator?",{"text":84,"@type":76},"The forced-response operator is constructed to be monotone in the control input, and it is identified from one to three days of control excitation. Together with the free response, it forms a counterfactual-capable world model for MPC.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]