Reinforcement Learning for Building Environmental Control

Dalamagkidis, Konstantinos; Kolokotsa, Dionysia

doi:10.5772/5286

Cited by 17 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, Dalamagkidis and Kolokotsa (2008) studied the overall "environmental control" of buildings and emphasized the importance of the implementation of "intelligent control techniques" to monitor all the factors responsible for a building's sustainability in terms of thermal comfort, visual comfort, air quality, and energy conservation. To this same purpose, the research by Morello et al (2009) demonstrated the important role that urban design and urban configurations play in the performance of individual buildings.…”

Section: Background and Objectivesmentioning

confidence: 99%

Simulating the Thermal-Energy Performance of Buildings at the Urban Scale: Evaluation of Inter-Building Effects in Different Urban Configurations

Pisello¹,

Castaldo²,

Poli³

et al. 2014

Journal of Urban Technology

View full text Add to dashboard Cite

The reduction of the energy use of buildings at the urban scale represents a key research and design topic with the purpose of developing specific methods for saving energy in buildings. These methods are often focused on the analysis of building thermal-energy behavior by considering the building as a stand-alone object. In this perspective, the thermal-energy behavior of two buildings in New York City is evaluated in this paper with varying urban contexts, in order to evaluate the Inter-Building Effect (IBE). The IBE analysis shows that the inaccuracy in neglecting the urban context in building modeling varies from 9.6 percent, to 71.9 percent. These results confirm that, in order to make accurate predictions about building thermal-energy performance in real urban contexts through dynamic simulation, the interaction between the building and the urban surrounding should be taken into account, in particular in dense urban areas.

show abstract

Section: Background and Objectivesmentioning

confidence: 99%

Simulating the Thermal-Energy Performance of Buildings at the Urban Scale: Evaluation of Inter-Building Effects in Different Urban Configurations

Pisello¹,

Castaldo²,

Poli³

et al. 2014

Journal of Urban Technology

View full text Add to dashboard Cite

show abstract

“…A couple of papers by Dalamagkidis and Kolokotsa [43,44] present an implementation of RL using a temporal difference (TD) error method for optimising the thermal comfort control of commercial buildings without requiring an environment model. In these works, the authors coupled the RL with a recursive least squares method to increase the method convergence speed.…”

Section: Applications Of Reinforced Learning To Energy Systemsmentioning

confidence: 99%

“…Both papers provide details on how the state space is constructed and employ a stochastic framework for the problem formulation. Yang et al [44] is very effective in describing the problem, the demo case, the theoretical references, and the details of the implementations. Moreover, it deals with the curse of dimensionality by coupling Q-learning with artificial neural networks.…”

Section: Applications Of Reinforced Learning To Energy Systemsmentioning

confidence: 99%

“…Moreover, it deals with the curse of dimensionality by coupling Q-learning with artificial neural networks. In Yang et al [44], the benefits are a result of the combination of reinforced learning and the solution for an optimization problem, while Kim and Lim [45] solve the optimization problem implicitly using the Q-learning algorithm itself. Both papers are relevant to this work, although here the scope is narrower: not the building but the solar DHW system alone.…”

Section: Applications Of Reinforced Learning To Energy Systemsmentioning

confidence: 99%

See 1 more Smart Citation

Analysis and Adaptation of Q-Learning Algorithm to Expert Controls of a Solar Domestic Hot Water System

Bettoni

Soppelsa

Fedrizzi

et al. 2019

ASI

View full text Add to dashboard Cite

This paper discusses the development of a coupled Q-learning/fuzzy control algorithm to be applied to the control of solar domestic hot water systems. The controller brings the benefit of showing performance in line with the best reference controllers without the need for devoting time to modelling and simulations to tune its parameters before deployment. The performance of the proposed control algorithm was analysed in detail concerning the input membership function defining the fuzzy controller. The algorithm was compared to four standard reference control cases using three performance figures: the seasonal performance factor of the solar collectors, the seasonal performance factor of the system and the number of on/off cycles of the primary circulator. The work shows that the reinforced learning controller can find the best performing fuzzy controller within a family of controllers. It also shows how to increase the speed of the learning process by loading the controller with partial pre-existing information. The new controller performed significantly better than the best reference case with regard to the collectors’ performance factor (between 15% and 115%), and at the same time, to the number of on/off cycles of the primary circulator (1.2 per day down from 30 per day). Regarding the domestic hot water performance factor, the new controller performed about 11% worse than the best reference controller but greatly improved its on/off cycle figure (425 from 11,046). The decrease in performance was due to the choice of reward function, which was not selected for that purpose and it was blind to some of the factors influencing the system performance factor.

show abstract

“…Various building technology studies have been published, such as occupant behavior (Hong et al 2016;O'Neill and Niu 2017), energy conservation measures (Qian et al 2018), sensors and controls (Li and Wen 2014), etc. Among them, controls were proven to be effective in reducing energy consumption (Dalamagkidis and Kolokotsa 2008;Lau et al 2014). Control loops are widely used in the process industry and HVAC systems.…”

Section: Introductionmentioning

confidence: 99%

Development of control quality factor for HVAC control loop performance assessment—II: Field testing and results (ASHRAE RP-1587)

Liu

O’Neill

et al. 2019

Science and Technology for the Built Environment

View full text Add to dashboard Cite

This article is the third paper from the research project RP-1587, focusing on presenting a comprehensive field test of the proposed control quality factors (CQFs; i.e., CQF-Harris and CQF-exponentially weighted moving average [EWMA]) and testing results. Firstly, the simulated control loops and real control loops are evaluated for offline assessment. Then, the field experiment implemented for different HVAC control loops is assessed online using the proposed CQFs. Test results show that the proposed CQFs are capable of adequately and effectively assessing the HVAC control loop performance. The methodology of obtaining those CQFs is provided in the companion paper: Development of Control Quality Factor for HVAC Control Loop Performance Assessment I-Methodology (ASHRAE RP-1587) (Li et al. 2019).

show abstract

Reinforcement Learning for Building Environmental Control

Cited by 17 publications

References 27 publications

Simulating the Thermal-Energy Performance of Buildings at the Urban Scale: Evaluation of Inter-Building Effects in Different Urban Configurations

Simulating the Thermal-Energy Performance of Buildings at the Urban Scale: Evaluation of Inter-Building Effects in Different Urban Configurations

Analysis and Adaptation of Q-Learning Algorithm to Expert Controls of a Solar Domestic Hot Water System

Development of control quality factor for HVAC control loop performance assessment—II: Field testing and results (ASHRAE RP-1587)

Contact Info

Product

Resources

About