A comprehensive review of the applications of machine learning for HVAC

Zhou, S.L.; Shah, A.A.; Leung, P.K.; Zhu, X.; Liao, Q.

doi:10.1016/j.decarb.2023.100023

Cited by 26 publications

(4 citation statements)

References 172 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on this state, the agent selects an action according to its policy. The policy is essentially the strategy or behavior that the agent follows, and it can be deterministic (always giving the same action for a given state) or stochastic (providing a distribution over possible actions) [23,40].…”

Section: The General Reinforcement Learning Conceptual Backgroundmentioning

confidence: 99%

“…To aid in this learning, the agent often estimates a value function, which predicts the expected return from a given state when following a particular policy. This value function helps the agent to judge the long-term consequences of its actions, enabling it to favor actions that lead to higher cumulative rewards in the future [23,40]. More specifically:…”

Section: The General Reinforcement Learning Conceptual Backgroundmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Reinforcement Learning Algorithms in Residential Energy Saving and Comfort Management

Lazaridis,

Michailidis,

Karatzinis

et al. 2024

Energies

View full text Add to dashboard Cite

The challenge of maintaining optimal comfort in residents while minimizing energy consumption has long been a focal point for researchers and practitioners. As technology advances, reinforcement learning (RL)—a branch of machine learning where algorithms learn by interacting with the environment—has emerged as a prominent solution to this challenge. However, the modern literature exhibits a plethora of RL methodologies, rendering the selection of the most suitable one a significant challenge. This work focuses on evaluating various RL methodologies for saving energy while maintaining adequate comfort levels in a residential setting. Five prominent RL algorithms—Proximal Policy Optimization (PPO), Deep Deterministic Policy Gradient (DDPG), Deep Q-Network (DQN), Advantage Actor-Critic (A2C), and Soft Actor-Critic (SAC)—are being thoroughly compared towards a baseline conventional control approach, exhibiting their potential to improve energy use while ensuring a comfortable living environment. The integrated comparison between the different RL methodologies emphasizes the subtle strengths and weaknesses of each algorithm, indicating that the best selection relies heavily on particular energy and comfort objectives.

show abstract

Section: The General Reinforcement Learning Conceptual Backgroundmentioning

confidence: 99%

Section: The General Reinforcement Learning Conceptual Backgroundmentioning

confidence: 99%

Evaluating Reinforcement Learning Algorithms in Residential Energy Saving and Comfort Management

Lazaridis,

Michailidis,

Karatzinis

et al. 2024

Energies

View full text Add to dashboard Cite

show abstract

“…To diminish energy demand within this OPEN ACCESS EDITED BY Amirhossein Balali, The University of Manchester, United Kingdom sector, building designers have various solutions at their disposal. These include implementing energy-saving technologies to enhance the thermal performance of the building envelopes (Berardi and Naldi, 2017;Huang et al, 2021), optimizing heating, ventilation, and air conditioning (HVAC) equipment efficiency (Conceição et al, 2021;Zhou et al, 2023), and integration renewable energy sources (RES) (Christopher et al, 2023;Lebedeva et al, 2023). Recently, the significance of building automation and control systems (BACS) in decreasing the energy demand of buildings has been increasingly acknowledged (Vandenbogaerde et al, 2023).…”

Section: Introductionmentioning

confidence: 99%

Optimizing office building operations: a framework for continuous dynamic energy simulations in decision-making for efficiency

Borodinecs,

Palcikovskis,

Krumins

et al. 2024

Front. Built Environ.

View full text Add to dashboard Cite

Digital twins represent a promising approach for sustainable building operations and management in the context of the carbon neutrality goals of the European Union (EU). Using OpenStudio, an opensource platform for building energy modeling, we demonstrated the creation and editing of building digital twins. OpenStudio provides a user-friendly interface and extensive simulation capabilities, allowing detailed and accurate modeling of building components and systems. Using OpenStudio Measures, users can automate tasks and customize simulation models to optimize the building performance. The process of creating a building digital twin involves collecting historical data and accurately representing the building geometry; materials; schedules; and heating, ventilation, and air conditioning (HVAC) systems. Challenges such as data availability and model accuracy highlight the importance of modeling practices. Editing the digital twin involves modifying the OpenStudio model files and EnergyPlus weather files to simulate different building operation scenarios. Python programming language opportunities were considered for digital twin file modification. The potential of digital twins lies in their ability to simulate future building conditions and optimize building system settings. By integrating digital twins with machine learning algorithms and connecting them directly to building management systems, optimal building control strategies can be automated, thereby reducing energy consumption and improving occupant comfort levels.

show abstract

“…Thermal comfort prediction models offer an opportunity to respond to both individuals' comfort needs and energy efficiency [11]. Machine learning has great potential for improving prediction models for the thermal comfort and performance of building heating, ventilation, and air conditioning (HVAC) systems, aiding designers in developing energy savings, and indoor environmental quality [12][13][14][15].…”

Section: Introductionmentioning

confidence: 99%

Optimising Building Energy and Comfort Predictions with Intelligent Computational Model

Alghamdi,

Tang,

Kanjanabootra

et al. 2024

Sustainability

View full text Add to dashboard Cite

Building performance prediction is a significant area of research, due to its potential to enhance the efficiency of building energy management systems. Its importance is particularly evident when such predictions are validated against field data. This paper presents an intelligent computational model combining Monte Carlo analysis, Energy Plus, and an artificial neural network (ANN) to refine energy consumption and thermal comfort predictions. This model addresses various combinations of architectural building design parameters and their distributions, effectively managing the complex non-linear relationships between the response variables and predictors. The model’s strength is demonstrated through its alignment with R2 values exceeding 0.97 for both thermal discomfort hours and energy consumption during the training and testing phases. Validation with field investigation data further confirms its accuracy, demonstrating average relative errors below 2.0% for total energy consumption and below 1.0% for average thermal discomfort hours. In particular, an average underestimation of −12.5% in performance discrepancies is observed when comparing the building energy simulation model with field data, while the intelligent computational model presented a smaller overestimation error (of +8.65%) when validated against the field data. This discrepancy highlights the model’s potential and reliability for the simulation of real-world building performance metrics, marking it as a valuable tool for practitioners and researchers in the field of building sustainability.

show abstract

A comprehensive review of the applications of machine learning for HVAC

Cited by 26 publications

References 172 publications

Evaluating Reinforcement Learning Algorithms in Residential Energy Saving and Comfort Management

Evaluating Reinforcement Learning Algorithms in Residential Energy Saving and Comfort Management

Optimizing office building operations: a framework for continuous dynamic energy simulations in decision-making for efficiency

Optimising Building Energy and Comfort Predictions with Intelligent Computational Model

Contact Info

Product

Resources

About