Pith. sign in

REVIEW 4 major objections 5 minor 45 references

Human-in-the-Loop AI for HVAC Management Enhancing Comfort and Energy Efficiency

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims an HVAC controller can learn occupant comfort from overrides and respond to price signals, cutting energy cost without predefined comfort models.

desk verdict Competent HITL HVAC simulation with a plausible energy-cost story, but its own Table 1 contradicts the comfort claim and the comfort evaluation is self-referential; the energy and framework parts deserve peer review, the headline does not. read the letter →

arxiv 2505.05796 v1 pith:WE7ES3F2 submitted 2025-05-09 eess.SY cs.AIcs.SYmath.OC

Xinyu Liang , Frits de Nijs , Buser Say , Hao Wang This is my paper · ORCID
classification eess.SYcs.AIcs.SYmath.OC
keywords HVACcontrolhuman-in-the-loopreinforcementlearningoccupantcomfortfeedbackbufferoccupancypredictionelectricitypricedemandresponse
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that an HVAC controller can learn an occupant's comfort preferences from the occupant's own override actions, rather than from a predefined thermal-comfort model or a fixed setpoint schedule. The proposed human-in-the-loop controller is a reinforcement-learning agent that keeps a buffer of recent overrides, uses that buffer both to correct its immediate action and to define a discomfort cost, and simultaneously optimizes against forecasted wholesale electricity prices and predicted occupancy. In simulation, the authors report significant energy-cost reductions relative to a rule-based occupancy thermostat, with occupant comfort maintained or improved in their discomfort-cost accounting, and performance close to an optimization controller that is given perfect forecasts and explicit comfort constraints. Because HVAC accounts for roughly 38% of building energy use, a controller that personalizes comfort without sensors or manual programming while shifting load to cheap, low-emission hours would be a practical route to demand response.

What carries the argument

The central object is the feedback buffer, a fixed-length window of occupant override signals taking values in $\{-1,0,1\}$, included in the MDP state and updated each time step. It does three jobs at once: it immediately corrects the applied control action, it enters the state representation so the policy can learn which actions provoke overrides, and it defines the discomfort cost through an exponentially weighted sum of recent override events. The companion mechanism is the simulated feedback generator, which sets the override probability to $p^f_t = \min\left(\left((T_t^{\mathrm{in}} - T^{\mathrm{set}})/\theta_{\mathrm{range}}\right)^2, p_{\max}\right)$. This quadratic link between temperature deviation and override likelihood is what lets the framework claim that comfort preferences are learned from observed interventions rather than from a predefined comfort model.

What would settle it

Replace the simulated occupant in the feedback equations with real logged override data from a deployed thermostat, or with a different discomfort model (for example, one with hysteresis, asymmetric heating and cooling responses, or delayed feedback), and re-run the same PPO training and evaluation. If the feedback-driven policy no longer beats the rule-based baseline in total cost, or if override frequency does not fall as training proceeds, the claim that comfort preferences are genuinely learned from feedback rather than encoded in the simulator is refuted.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that occupant discomfort can be treated as a directly observable feedback signal instead of an unmeasured deviation from a comfort model, and that a policy trained on that signal can simultaneously track electricity prices. The agent's state includes the current indoor and outdoor temperatures, time of day, forecasted outdoor temperature, current and forecasted occupancy, a buffer of recent occupant overrides, and current and forecasted wholesale market rates. The action is simply whether the HVAC is on or off. The reward subtracts an exponentially weighted penalty for recent overrides from the cost of electricity consumed, and the policy is learned with PPO. The paper's simulation generates overrides from a capped quadratic function of the distance between indoor temperature and a comfort setpoint, and its experiments compare four levels of occupancy information, a rule-based thermostat, and a perfect-foresight optimization benchmark. The reported outcome is that the feedback-driven policy cuts energy cost relative to the rule-based controller, that occupancy forecasts help most when comfort is weighted heavily, and that the learned controller stays close to the optimization ideal despite having no explicit comfort constraints.

Load-bearing premise

The framework's simulation assumes an occupant's probability of overriding the controller is a capped quadratic function of how far indoor temperature is from one known setpoint, so both the reward that trains the agent and the comfort metrics used to evaluate it are generated from that same formula; if real occupants override for other reasons or with other patterns, the claimed automatic learning of comfort preferences is not supported.

Editorial extensions

If this is right

  • If the central claim holds, a thermostat could be replaced by a learned policy that treats override events as the only comfort signal, so occupants never need to program schedules or understand a comfort model.
  • Because wholesale rates and their forecasts are part of the state, the agent learns to pre-cool or pre-heat when energy is cheap and occupancy is expected, shifting HVAC load toward periods of grid surplus and supporting demand response.
  • The scenario comparison indicates that explicit occupancy information improves the cost–comfort trade-off, but the feedback buffer alone already provides enough signal to operate reasonably when occupancy data are missing or privacy-sensitive.
  • Sensitivity results across maximum feedback probabilities imply the learned policy remains serviceable when occupants do not always report discomfort, so imperfect human feedback does not by itself break the approach.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the simulation generates both the learning signal and the evaluation metrics from the same $T^{\mathrm{set}}$-anchored override formula, so the strongest validation would be a real deployment or a second simulator with a different occupant model; otherwise the claim that preferences are learned rather than assumed may be circular.
  • Editorial inference: the same feedback-buffer representation should transfer to other building systems with observable occupant overrides—lighting, window shading, or plug loads—where an explicit comfort model is expensive but a binary corrective signal is cheap.
  • Editorial inference: the paper's future-work mention of delayed feedback suggests a concrete test: if overrides are shifted by one or more time steps to mimic occupants who react late, the policy must still reduce override frequency and cost for the framework to hold in realistic settings.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a Human-in-the-Loop (HITL) reinforcement-learning framework for residential HVAC control. The controller observes indoor/outdoor temperatures, time-of-day features, forecasted outdoor temperature, occupancy and occupancy forecasts, a buffer of recent occupant overrides, and current/forecasted wholesale electricity prices, and chooses whether to run the HVAC (Secs. 2.2-2.3). Occupant discomfort is simulated as a stochastic override triggered with probability proportional to the squared deviation of indoor temperature from a setpoint T_set = 22°C (Eqs. 10-13), and the reward penalizes these overrides plus energy cost (Eqs. 16-20). PPO is used for learning, with a bidirectional-LSTM occupancy predictor for the realistic scenario (Sec. 3). Experiments over 30 days of May 2023 Melbourne data compare four occupancy-information scenarios against a rule-based controller and a perfect-information optimization controller (Secs. 4-5). The headline claim is that the method cuts energy cost while maintaining or enhancing occupant comfort.

Significance. If the results held, the framework would be a useful step toward demand-response-aware HVAC control that learns occupant preferences from lightweight interaction signals rather than from calibrated thermal-comfort models. The manuscript has concrete strengths: a transparent MDP formulation with an explicit thermal model (Eq. 8), reproducible data sources (ARAS, Visual Crossing, AEMO), a sensitivity analysis of the feedback probability cap (Fig. 3), and an explicit comparison with a rolling-horizon optimization benchmark. However, the evaluation as presented does not support the comfort-preservation claim: the reported comfort metrics are worse than the rule-based baseline, and the comfort reward/evaluation loop is generated from the same setpoint the method is supposed to learn. These issues are load-bearing for the central contribution.

major comments (4)
  1. [Table 1 and Sec. 5.3] The abstract and Sec. 1 contribution (4) claim the method 'maintains or enhances occupant comfort' and achieves savings 'while maintaining or enhancing occupant comfort when compared to rule-based strategies.' Table 1 directly contradicts this: HITL has a 10.24% temperature violation probability and 1.82°C MAE to setpoint, versus 3.13% and 0.49°C for the rule-based controller and 0.00% violation probability for the optimization controller. Because the table averages over all scenarios and discomfort proportions beta, it is possible that some beta configuration preserves comfort, but the paper does not report such a configuration. The authors should either report a disaggregated table identifying a scenario/beta where HITL meets or improves on the rule-based comfort metrics, or revise the abstract and contribution statements to claim cost savings only.
  2. [Secs. 2.4.2, 2.5.1, and 5] The comfort component is self-referential. The simulated feedback probability p_f depends on T_set in Eq. (10), the expected comfort-preserving action a* depends on T_set in Eq. (11), the discomfort reward in Eq. (16) is computed from the resulting feedback buffer, and the evaluation metrics (violation probability and MAE to setpoint) in Sec. 5 are both defined relative to the same T_set = 22°C. Consequently, the agent is rewarded and evaluated against the very preference that generates its feedback signal, which makes the claim in Sec. 2.5.1 that the model 'dynamically adjusts based on the feedback that directly reflects occupant comfort' circular. To support the stated contribution of learning preferences without predefined comfort models, the authors need at least one evaluation against a comfort measure that is not derived from T_set (e.g., a separate occupant model or field data), or they must explicitly frame the contribution as tuning to a known setpoint.
  3. [Sec. 5.1 and Fig. 2(c)] The section claims the system learns preferences with 'limited need for user intervention' and that S2 achieves 'a reasonable number of overrides,' but no aggregate override counts or frequencies are reported. Fig. 2(c) shows only a four-day excerpt for S1, S2, S4, and no statistics over the 25 runs or across the seven test days, and there is no comparison with the override rate of the rule-based or optimization controllers. The authors should report quantitative override rates with standard deviations, which is the natural metric for evaluating the claimed reduction in manual interaction.
  4. [Secs. 4.4.2 and 5.3] The optimization benchmark holds indoor temperature within the comfort range as a hard constraint, so its 0% violation probability is guaranteed by construction. Comparing HITL's 10.24% violation probability with this infeasible benchmark does not support the claim in Sec. 1 contribution (4) that HITL 'achieves performance close to the theoretical ceiling.' A more informative comparison would report the optimization controller's energy cost either with the same discomfort weight beta in the objective or with a softened comfort constraint, so the cost-comfort trade-off can be assessed fairly.
minor comments (5)
  1. [Sec. 4.2] The text states that 'longer horizons were considered' but provides no experiment supporting the diminishing-returns claim; the authors should either show the comparison or remove the statement.
  2. [Table 1] Table 1 reports averages over 25 runs without confidence intervals or standard deviations; given the stochasticity noted in Sec. 5, the authors should provide dispersion measures or box plots for the temperature metrics.
  3. [Sec. 4.4.1] The rule-based controller description does not state the setpoint used; given that Table 1 reports MAE to setpoint, the setpoint should be specified for reproducibility.
  4. [Figs. 2(b) and 2(c)] These figures omit Scenario 3 and the rule-based controller 'due to page limitations'; since S3 is one of the four central scenarios, the figures should be moved to an appendix or the exclusion criterion should be stated.
  5. [Eq. (33)] Equation (33) is described as calculated using 'observable heating and cooling behavior,' but the values T_lower, T_upper, and t_cool are assumed rather than measured in this study; the wording should be adjusted to 'estimated' or 'assumed.'

Circularity Check

1 steps flagged · score 4.0 of 10

Comfort evaluation and the simulated feedback loop share the same fixed T_set, making the comfort claim partly self-referential; the energy-cost result remains externally grounded.

  1. self definitional [Section 2.4.2 (Eqs. 10-13), Section 2.5.1 (Eq. 16), and Section 5 (Table 1 metrics)]
    "Unlike predefined setpoints used in other system, in our framework these feedback signals are crucial for guiding the agent’s decision-making. At each time step, the feedback probability, p f t , is calculated to determine whether an occupant would override the current system due to discomfort. This probability is influenced by the difference between the current indoor temperature, T in t , and a set comfort temperature, T set. ... Mean Absolute Error (MAE) to setpoint temperature (i.e."

    The only comfort signal available to the agent is the simulated override f_t, which is generated in Eqs. (10)-(13) from a fixed hidden T_set = 22°C and θ_range = 3. The discomfort reward in Eq. (16) is computed from that same f_t, so the agent is trained to satisfy a comfort model that is predefined by T_set. The paper then validates 'maintaining or enhancing occupant comfort' with violation probability and MAE measured against the same T_set = 22°C and the same ±3°C comfort range. Therefore the learned preference and the reported comfort metric are two expressions of the same input constant, so the comfort result is not independently confirmed. The energy component uses external AEMO price and weather data, so the circularity is only partial.

full rationale

The main derivation chain — MDP formulation, thermal dynamics, PPO policy optimization, occupancy prediction, and energy-cost calculation from external price data — is self-contained and does not depend on author self-citation or an imported uniqueness theorem. The one genuinely self-referential element is the comfort loop: the simulated occupant feedback that defines the discomfort reward is generated from T_set, and the comfort evaluation metrics are also defined from the same T_set and comfort range. Because the agent never observes T_set directly, the setup is a legitimate hidden-parameter RL problem rather than a fully circular derivation; however, the claim that the framework works 'without predefined settings' is weakened by Eq. (10), which is itself a predefined setpoint model. It is also worth noting, as a correctness concern rather than circularity, that the paper's own Table 1 contradicts the abstract: HITL's violation probability (10.24%) and MAE (1.82°C) are worse than the rule-based baseline (3.13%, 0.49°C), so the comfort claim is not merely unverified but internally unsupported. Overall, the core energy optimization is externally grounded, giving a partial circularity score of 4.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

The central claim depends on several hand-chosen simulation parameters (T_set, theta_range, p_max, feedback weights, beta) and on the assumption that the synthetic occupant model reflects real behavior. The energy and weather data are external, but the comfort loop is defined internally and evaluated against the same setpoint.

free parameters (7)
  • Comfort setpoint T_set = 22 degrees Celsius
    Set from productivity meta-analysis [33]; used to generate simulated feedback (Eq. 10) and to compute MAE in evaluation, so it drives both learning and the reported comfort metric.
  • Comfort range theta_range = 3 degrees Celsius
    Set to ASHRAE Standard 55 range; enters the quadratic feedback probability in Eq. (10).
  • Maximum feedback probability p_max = 1.0 during training; swept 0.5-1.0 in sensitivity analysis
    Caps the Bernoulli override probability in Eq. (10) and is varied in Section 5.4.
  • Feedback buffer horizon h_f = 16 steps (4 hours)
    Chosen by hand in Eq. (14) and used in the discomfort cost weighting Eq. (16)-(17).
  • Discomfort weight shape e^{-e i / h_f} = Not a single value
    Hand-designed exponential weighting of recent feedback; not derived from data.
  • Reward weighting beta = Swept 0.1 to 0.9
    Balances discomfort versus energy in Eq. (19); not fitted but controls all reported trade-offs.
  • Occupied no-feedback penalty epsilon = Not specified
    Small negative reward in Eq. (16); the actual value is never stated, making the reward scale ambiguous.
assumptions (6)
  • standard math MDP framework and PPO clipped surrogate objective are standard and applicable.
    Assumed background for the learning formulation in Sections 2 and 3.2.
  • domain assumption Thermal model Eq. (8) from Mortensen and Haggerty [25] with RC parameters is a valid description of building thermal dynamics.
    The temperature transition and all cost comparisons rely on this linear RC model.
  • ad hoc to paper Occupant override behavior follows the quadratic Bernoulli model in Eqs. (10)-(13).
    This is the paper's synthetic human model; no real occupant data are used to validate it.
  • domain assumption ARAS 'Going Out' activity labels correctly represent true absence for occupancy.
    The entire occupancy input and the occupancy predictor are built on this label as ground truth.
  • domain assumption Wholesale price and weather forecasts are available as inputs to the controller.
    The state includes forecasted values as if provided by external services; no forecasting error model is included except for occupancy.
  • domain assumption Comfort is adequately measured by deviation from a single fixed setpoint 22 degrees Celsius with plus or minus 3 degrees Celsius range.
    The comfort metrics in Table 1 and the discomfort cost both assume this single-setpoint definition, which may not represent varied occupant preferences.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Human-in-the-Loop AI for HVAC Management Enhancing Comfort and Energy Efficiency." pith.science (2026). https://pith.science/paper/WE7ES3F2

@misc{pith2026250505796,
  author       = {Pith},
  title        = {Pith review of: Human-in-the-Loop AI for HVAC Management Enhancing Comfort and Energy Efficiency},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WE7ES3F2}},
  note         = {Machine review of arXiv:2505.05796}
}
read the original abstract

Heating, Ventilation, and Air Conditioning (HVAC) systems account for approximately 38% of building energy consumption globally, making them one of the most energy-intensive services. The increasing emphasis on energy efficiency and sustainability, combined with the need for enhanced occupant comfort, presents a significant challenge for traditional HVAC systems. These systems often fail to dynamically adjust to real-time changes in electricity market rates or individual comfort preferences, leading to increased energy costs and reduced comfort. In response, we propose a Human-in-the-Loop (HITL) Artificial Intelligence framework that optimizes HVAC performance by incorporating real-time user feedback and responding to fluctuating electricity prices. Unlike conventional systems that require predefined information about occupancy or comfort levels, our approach learns and adapts based on ongoing user input. By integrating the occupancy prediction model with reinforcement learning, the system improves operational efficiency and reduces energy costs in line with electricity market dynamics, thereby contributing to demand response initiatives. Through simulations, we demonstrate that our method achieves significant cost reductions compared to baseline approaches while maintaining or enhancing occupant comfort. This feedback-driven approach ensures personalized comfort control without the need for predefined settings, offering a scalable solution that balances individual preferences with economic and environmental goals.

Figures

Figures reproduced from arXiv: 2505.05796 by the authors.

Figure 1
Figure 1. Overview of HITL Energy Management System. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Comparison of different control strategies in terms of objectives, temperatures, and control decisions plus feedback. [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Sensitivity analysis of the HITL framework under different maximum probability caps 𝑝max with discomfort proportion 𝛽 = 0.5. robust performance through adaptation to feedback and market conditions. 5.4 Sensitivity to Feedback Responsiveness [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 43 canonical work pages

  1. [1]

    Moatassem Abdallah, Caroline Clevenger, Tam Vu, and Anh Nguyen. 2016. Sens- ing occupant comfort using wearable technologies. In Construction Research Congress 2016. American Society of Civil Engineers (ASCE), Reston, VA, 940–950

    work page 2016

  2. [2]

    Hande Alemdar, Halil Ertan, Ozlem Durmaz Incel, and Cem Ersoy. 2013. ARAS human activity datasets in multiple homes with multiple residents. In 2013 7th International Conference on Pervasive Computing Technologies for Healthcare and Workshops. IEEE, Piscataway, NJ, USA, 232–235

    work page 2013

  3. [3]

    Abdullah Alsalemi, Mona Ramadan, Faycal Bensaali, Abbes Amira, Christos Sar- dianos, Iraklis Varlamis, and George Dimitrakopoulos. 2019. Endorsing domestic energy saving behavior using micro-moment classification. Applied Energy 250 (2019), 1302–1311

    work page 2019

  4. [4]

    Kadir Amasyali and Nora M El-Gohary. 2016. Energy-related values and satisfac- tion levels of residential and office building occupants. Building and Environment 95 (2016), 251–263

    work page 2016

  5. [5]

    Uzma Amin, MJ Hossain, and E Fernandez. 2020. Optimal price based control of HVAC systems in multizone office buildings for demand response. Journal of Cleaner Production 270 (2020), 122059

    work page 2020

  6. [6]

    ANSI Ashrae et al. 2020. Standard 55-2020: Thermal environmental conditions for human occupancy . American Society of Heating, Refrigerating, and Air- Conditioning Engineers, Inc. Atlanta

    work page 2020

  7. [7]

    Australian Energy Market Operator (AEMO). 2024. Aggregated Price and Demand Data. https://aemo.com.au/en/energy-systems/electricity/national-electricity- market-nem/data-nem/aggregated-data. Accessed: 2024-01-15

    work page 2024

  8. [8]

    Mesut Avci, Murat Erkoc, Amir Rahmani, and Shihab Asfour. 2013. Model predictive HVAC load control in buildings using real-time electricity pricing. Energy and Buildings 60 (2013), 199–209

    work page 2013

Show all 45 references
  1. [9]

    Shahab Bahrami, M Hadi Amini, Miadreza Shafie-khah, and Joao PS Catalao

  2. [10]

    Liangliang Chen, Fei Meng, and Ying Zhang. 2023. Fast human-in-the-loop control for hvac systems via meta-learning and model-based offline reinforcement learning. IEEE Transactions on Sustainable Computing 8, 3 (2023), 504–521

    2023

  3. [11]

    Xiaogang Cheng, Bin Yang, Thomas Olofsson, Guoqing Liu, and Haibo Li. 2017. A pilot study of online non-invasive measuring technology based on video mag- nification to determine skin temperature. Building and Environment 121 (2017), 1–10

    2017

  4. [12]

    Adrian Chojecki, Michał Rodak, Arkadiusz Ambroziak, and Piotr Borkowski

  5. [13]

    Panos Constantopoulos, Fred C Schweppe, and Richard C Larson. 1991. ESTIA: A real-time consumer control scheme for space conditioning usage under spot electricity pricing. Computers & operations research 18, 8 (1991), 751–765

    1991

  6. [14]

    Zsuzsanna Csereklyei, Songze Qu, and Tihomir Ancev. 2019. The effect of wind and solar power generation on wholesale electricity prices in Australia. Energy Policy 131 (2019), 358–369

    2019

  7. [15]

    M González-Torres, Luis Pérez-Lombard, Juan F Coronel, Ismael R Maestre, and Da Yan. 2022. A review on buildings energy information: Trends, end-uses, fuels and drivers. Energy Reports 8 (2022), 626–637

    2022

  8. [16]

    Mengjie Han, Ross May, Xingxing Zhang, Xinru Wang, Song Pan, Da Yan, Yuan Jin, and Liguo Xu. 2019. A review of reinforcement learning methodologies for controlling occupant comfort in buildings. Sustainable cities and society 51 (2019), 101748

    2019

  9. [17]

    Farrokh Jazizadeh, Ali Ghahramani, Burcin Becerik-Gerber, Tatiana Kichkaylo, and Michael Orosz. 2014. Human-building interaction framework for personal- ized thermal comfort-driven systems in office buildings. Journal of Computing in Civil Engineering 28, 1 (2014), 2–16

    2014

  10. [18]

    Farrokh Jazizadeh and S. Pradeep. 2016. Can computers visually quantify hu- man thermal comfort? Short Paper. In Proceedings of the 3rd ACM International Conference on Systems for Energy-Efficient Built Environments (Palo Alto, CA, USA) (BuildSys ’16). Association for Computin...

    2016

  11. [19]

    Camille John, Charalampos Vallianos, José Candanedo, and Andreas Athienitis

  12. [20]

    Wooyoung Jung and Farrokh Jazizadeh. 2018. Vision-based thermal comfort quantification for HVAC control. Building and Environment 142 (2018), 513–523

    2018

  13. [21]

    Pushpendu Kar, Arish Shareef, Arun Kumar, Koh Tsyr Harn, Balaji Kalluri, and Sanjib Kumar Panda. 2019. ReViCEE: A recommendation based approach for personalized control, visual comfort & energy efficiency in buildings. Building and Environment 152 (2019), 135–144

    2019

  14. [22]

    Sami Karjalainen. 2009. Thermal comfort and use of thermostats in Finnish homes and offices. Building and Environment 44, 6 (2009), 1237–1245

    2009

  15. [23]

    Georgios I Maniatis and Nikolaos T Milonas. 2022. The impact of wind and solar power generation on the level and volatility of wholesale electricity prices in Human-in-the-Loop AI for HVAC Management Enhancing Comfort and Energy Efficiency E-ENERGY ’25, June 17–20, 2025, Rotte...

    2022

  16. [24]

    Amin Mirakhorli and Bing Dong. 2016. Occupancy behavior based model predict- ive control for building indoor climate—A critical review. Energy and Buildings 129 (2016), 499–513

    2016

  17. [25]

    Richard E Mortensen and Kevin P Haggerty. 1988. A stochastic computer model for heating and cooling loads. IEEE Transactions on Power Systems 3, 3 (1988), 1213–1219

    1988

  18. [26]

    Dan Popa, Florin Pop, Cristina Serbanescu, and Aniello Castiglione. 2019. Deep learning model for home automation and energy reduction in a smart home environment platform. Neural Computing & Applications 31, 5 (2019), 1317–1337

    2019

  19. [27]

    Marco Pritoni, Jonathan M Woolley, and Mark P Modera. 2016. Do occupancy- responsive learning thermostats save energy? A field study in university residence halls. Energy and Buildings 127 (2016), 469–478

    2016

  20. [28]

    M. L. Puterman. 2009. Markov Decision Processes: Discrete Stochastic Dynamic Programming. Wiley, Hoboken, NJ

    2009

  21. [29]

    Juhi Ranjan and James Scott. 2016. ThermalSense: determining dynamic thermal comfort preferences using thermographic imaging. InProceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing (Heidelberg, Germany) (UbiComp ’16). Association for...

    2016

  22. [30]

    Christos Sardianos, Christos Chronis, Iraklis Varlamis, George Dimitrakopoulos, Yassine Himeur, Abdullah Alsalemi, Faycal Bensaali, and Abbes Amira. 2020. Real-time personalised energy saving recommendations. In 2020 International conferences on internet of things (IThings) an...

    2020

  23. [31]

    Christos Sardianos, Iraklis Varlamis, George Dimitrakopoulos, Dimosthenis Anagnostopoulos, Abdullah Alsalemi, Faycal Bensaali, Yassine Himeur, and Abbes Amira. 2020. Rehab-c: Recommendations for energy habits change. Future Gen- eration Computer Systems 112 (2020), 394–407

    2020

  24. [32]

    John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov

  25. [33]

    Olli Seppanen, William J Fisk, and QH Lei. 2006. Room temperature and pro- ductivity in office work. Healthy Buildings 2006 Conference 1 (2006), 243–247

    2006

  26. [34]

    Salman Sadiq Shuvo and Yasin Yilmaz. 2022. Home energy recommendation sys- tem (hers): A deep reinforcement learning method based on residents’ feedback and activity. IEEE Transactions on Smart Grid 13, 4 (2022), 2812–2821

    2022

  27. [35]

    Energy Information Administration

    U.S. Energy Information Administration. 2018. Table HC6.1: Space heating in U.S. homes by housing unit type, 2015. https://www.eia.gov/consumption/ residential/data/2015/hc/php/hc6.1.php Accessed: 2024-09-24

    2018

  28. [36]

    arXiv preprint arXiv:1707.06347

    Proximal Policy Optimization Algorithms. arXiv preprint arXiv:1707.06347

    arXiv

  29. [37]

    Yixuan Wei, Xingxing Zhang, Yong Shi, Liang Xia, Song Pan, Jinshun Wu, Mengjie Han, and Xiaoyun Zhao. 2018. A review of data-driven approaches for prediction and classification of building energy consumption. Renewable and Sustainable Energy Reviews 82 (2018), 1027–1047

    2018

  30. [38]

    Ye Yao and Divyanshu Kumar Shekhar. 2021. State of the art review on model predictive control (MPC) in Heating Ventilation and Air-conditioning (HVAC) field. Building and Environment 200 (2021), 107952

    2021

  31. [39]

    Bo Yi and Joon-Ho Choi. 2015. Facial skin temperature as a proactive variable in a building thermal comfort control system. In Sustainable Human–Building Ecosystems. American Society of Civil Engineers, Reston, VA, 117–125

    2015

  32. [40]

    Visual Crossing Corporation. 2024. Weather Data Services. https://www. visualcrossing.com/weather/weather-data-services. Accessed: 2024-01-15

    2024

  33. [41]

    SL Zhou, AA Shah, PK Leung, X Zhu, and Q Liao. 2023. A comprehensive review of the applications of machine learning for HVAC. DeCarbon 2 (2023), 100023

    2023

  34. [44]

    Liang Yu, Di Xie, Chongxin Huang, Tao Jiang, and Yulong Zou. 2018. Energy optimization of HVAC systems in commercial buildings considering indoor air quality management. IEEE Transactions on Smart Grid 10, 5 (2018), 5103–5113

    2018

  35. [2017]

    IEEE Transactions on Power Systems 33, 4 (2017), 4218–4227

    A decentralized electricity market scheme enabling demand response deployment. IEEE Transactions on Power Systems 33, 4 (2017), 4218–4227

    2017

  36. [2018]

    In Proceed- ings of the 7th International Building Physics Conference

    Estimating time constants for over 10,000 residential buildings in North America: towards a statistical characterization of thermal dynamics. In Proceed- ings of the 7th International Building Physics Conference. International Association of Building Physics, Syracuse, NY, USA...

  37. [2020]

    IET Smart Grid 3, 2 (2020), 254–266

    Energy management system for residential buildings based on fuzzy logic: design and implementation in smart-meter. IET Smart Grid 3, 2 (2020), 254–266

    2020

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.