Toward Smart Multizone HVAC Control by Combining Context-Aware System and Deep Reinforcement Learning

Deng, Xiangtian; Zhang, Yi; He, Qi

doi:10.1109/jiot.2022.3175728

Cited by 13 publications

(6 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although in OCFP problem (6) only indoor average temperature is considered, a building is composed of a set of rooms, each one is equipped with a thermostat and indoor sensor. The sample input and output vectors collected for LSTM training at time step s are defined as follows: X(s) = y (1) , . .…”

Section: B Lstm-based Plant Thermodynamic Modelmentioning

confidence: 99%

“…At step 1, the system takes the followed class k j (h) as reference. At step 2, the State Observer collects current indoor temperatures: T (1) (s), . .…”

Section: Model Predictive Control Schemementioning

confidence: 99%

“…The non-linear problem can be solved by iterative methods, such as Sequential Least SQuares Programming (SLQP) [42]. Moreover, for each time step s, the Optimizer calculates a candidate optimal input vector X(s) for the LSTM model as follows: X(s) = y (1) (s + i), . .…”

Section: Model Predictive Control Schemementioning

confidence: 99%

“…T HE quest for energy efficiency and consumption control is fundamental from an environmental perspective. Fifty percent of building energy consumption is used in Heating, Ventilation, and Air-Conditioning (HVAC) systems [1]. Hence, effective HVAC control techniques for energy consumption minimization and thermal comfort guarantee have attracted the attention of researchers.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A User Based HVAC System Management Through Blockchain Technology and Model Predictive Control

Olivieri,

Volpe,

Mangini

et al. 2024

IEEE Trans. Automat. Sci. Eng.

View full text Add to dashboard Cite

This paper introduces an innovative approach to designing a user-based Heating, Ventilation, and Air-Conditioning (HVAC) system management connected with the District Energy Management System. By classifying the users into dynamic energy consumption classes to reward energy efficiency and penalize excessive use, users can modify their behavior to pass to a less expensive and more virtuous consumption class. To this aim, a blockchain platform determines the rewards and penalties and, by a K-means clustering algorithm, categorizes users into respective groups. Then, a Class Follower Problem is formulated and solved by a Model Predictive Control (MPC) strategy integrated with a Long Short-Term Memory network as a predictive model. If the users follow the suggestions proposed by the controller, i.e., the thermostat set-points and the time intervals in which the HVAC system must be switched off or on, the users can be located in a more virtuous consumption class. A case study conducted within an energy district in Bari (Italy) shows how the proposed architectural framework tuned thermal regulation in intelligent buildings while concurrently achieving energy optimization.Note to Practitioners-This paper addresses the challenge of efficiently managing HVAC systems in smart districts through a novel blockchain-based framework and an optimization strategy solved by an MPC approach. The objective is to incentivize users to optimize their energy consumption by introducing dynamic Consumption Classes that reward energy efficiency and penalize inefficient utilization. For practitioners, this strategy translates to a granular level of energy management that not only adapts to individual behaviors but also aligns with broader sustainability goals. Integrating the blockchain platform ensures a transparent and secure method for managing and recording energy usage. At the same time, adopting MPC with Long Short-Term Memory Networks offers accurate forecasts and adjustments to enhance system responsiveness. Although the study focuses on HVAC systems, the principles may be extended to other energy-intensive applications, providing a comprehensive tool for energy management and user engagement in smart cities. Future research could integrate renewable energy sources and explore the implications of user-driven adjustments on the overall energy distribution and efficiency.

show abstract

Section: B Lstm-based Plant Thermodynamic Modelmentioning

confidence: 99%

“…At step 1, the system takes the followed class k j (h) as reference. At step 2, the State Observer collects current indoor temperatures: T (1) (s), . .…”

Section: Model Predictive Control Schemementioning

confidence: 99%

Section: Model Predictive Control Schemementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A User Based HVAC System Management Through Blockchain Technology and Model Predictive Control

Olivieri,

Volpe,

Mangini

et al. 2024

IEEE Trans. Automat. Sci. Eng.

View full text Add to dashboard Cite

show abstract

“…Besides RNNs, other architectures that address the temporal dependencies can be used. For example, Deng et al [42] proposed to use a transformer [43] with SAC to address non-stationary environments.…”

Section: B Rnn For Reinforcement Learning In the Energy Sectormentioning

confidence: 99%

Data Center HVAC Control Harnessing Flexibility Potential via Real-Time Pricing Cost Optimization Using Reinforcement Learning

Biemann

Gunkel

Scheller

et al. 2023

IEEE Internet Things J.

View full text Add to dashboard Cite

With increasing electricity prices, cost savings through load shifting are becoming increasingly important for energy end users. While dynamic pricing encourages customers to shift demand to low price periods, the non-stationary and highly volatile nature of electricity prices poses a significant challenge to energy management systems. In this paper, we investigate the flexibility potential of data centres by optimising heating, ventilation, and air conditioning systems with a general modelfree reinforcement learning approach. Since the soft actor-critic algorithm with feed-forward networks did not work satisfactorily in this scenario, we propose instead a parameterisation with a recurrent neural network architecture to successfully handle spot-market price data. The past is encoded into a hidden state, which provides a way to learn the temporal dependencies in the observations and highly volatile rewards. The proposed method is then evaluated in experiments on a simulated data centre. Considering real temperature and price signals over multiple years, the results show a cost reduction compared to a proportional, integral and derivative controller while maintaining the temperature of the data centre within the desired operating ranges. In this context, this work demonstrates an innovative and applicable reinforcement learning approach that incorporates complex economic objectives into agent decision-making. The proposed control method can be integrated into various Internet of things-based smart building solutions for energy management.

show abstract

A novel operation method for renewable building by combining distributed DC energy system and deep reinforcement learning

Deng,

Zhang,

Jiang

et al. 2024

Applied Energy

View full text Add to dashboard Cite

Toward Smart Multizone HVAC Control by Combining Context-Aware System and Deep Reinforcement Learning

Cited by 13 publications

References 59 publications

A User Based HVAC System Management Through Blockchain Technology and Model Predictive Control

A User Based HVAC System Management Through Blockchain Technology and Model Predictive Control

Data Center HVAC Control Harnessing Flexibility Potential via Real-Time Pricing Cost Optimization Using Reinforcement Learning

A novel operation method for renewable building by combining distributed DC energy system and deep reinforcement learning

Contact Info

Product

Resources

About