Reinforcement learning for demand response: A review of algorithms and modeling techniques

Applied Energy, Journal Year: 2018, Volume and Issue: 235, P. 1072 - 1089

Published: Nov. 17, 2018

Language: Английский

---

Reinforcement learning for building controls: The opportunities and challenges

Applied Energy, Journal Year: 2020, Volume and Issue: 269, P. 115036 - 115036

Published: May 12, 2020

Language: Английский

---

AI-big data analytics for building automation and management systems: a survey, actual challenges and future perspectives

Artificial Intelligence Review, Journal Year: 2022, Volume and Issue: 56(6), P. 4929 - 5021

Published: Oct. 15, 2022

In theory, building automation and management systems (BAMSs) can provide all the components functionalities required for analyzing operating buildings. However, in reality, these only ensure control of heating ventilation air conditioning system systems. Therefore, many other tasks are left to operator, e.g. evaluating buildings' performance, detecting abnormal energy consumption, identifying changes needed improve efficiency, ensuring security privacy end-users, etc. To that end, there has been a movement developing artificial intelligence (AI) big data analytic tools as they offer various new tailor-made solutions incredibly appropriate practical management. Typically, help operator (i) tons connected equipment data; and; (ii) making intelligent, efficient, on-time decisions performance. This paper presents comprehensive systematic survey on using AI-big analytics BAMSs. It covers AI-based tasks, load forecasting, water management, indoor environmental quality monitoring, occupancy detection, The first part this adopts well-designed taxonomy overview existing frameworks. A review is conducted about different aspects, including learning process, environment, computing platforms, application scenario. Moving on, critical discussion performed identify current challenges. second aims at providing reader with insights into real-world analytics. Thus, three case studies demonstrate use BAMSs presented, focusing anomaly detection residential office buildings performance optimization sports facilities. Lastly, future directions valuable recommendations identified reliability intelligent

Language: Английский

---

Applications of reinforcement learning in energy systems

Renewable and Sustainable Energy Reviews, Journal Year: 2020, Volume and Issue: 137, P. 110618 - 110618

Published: Dec. 9, 2020

Energy systems undergo major transitions to facilitate the large-scale penetration of renewable energy technologies and improve efficiencies, leading integration many sectors into system domain. As complexities in this domain increase, it becomes challenging control flows using existing techniques based on physical models. Moreover, although data-driven models, such as reinforcement learning (RL), have gained considerable attention fields, a direct shift RL is not feasible irrespective ongoing complexities. To end, top-down approach used understand behavior by reviewing current state art. We classified papers literature seven categories their area application. Subsequently, publications under each category were further examined relative problem diversity, technique employed, performance improvement (compared with other white gray box models), verification, reproducibility; articles reported 10–20% use RL. In most studies, however, deep state-of-the-art actor-critic methods (e.g., twin delayed deterministic policy gradient soft actor-critic) applied. This has remarkably hindered improvements problems related complex been considered. Approximately half Q-learning. Furthermore, despite availability historical data domain, batch algorithms exploited. Emerging multi-agent applications may be considered positive development that can enable management interactions among multiple parties. Most studies lack proper benchmarking compared model-based approaches or gray-box majority cover dispatch building management. Although adequately solve are considerably integrated several sectors, only limited number discussed its broad The present study clearly demonstrates even without full utilization capacity, potential resolving continuously increasing complexity within

Language: Английский

---

Whole building energy model for HVAC optimal control: A practical framework based on deep reinforcement learning

Energy and Buildings, Journal Year: 2019, Volume and Issue: 199, P. 472 - 490

Published: July 16, 2019

Language: Английский

---

A review of reinforcement learning for autonomous building energy management

Computers & Electrical Engineering, Journal Year: 2019, Volume and Issue: 78, P. 300 - 312

Published: Aug. 2, 2019

Language: Английский

---

An overview of machine learning applications for smart buildings

Sustainable Cities and Society, Journal Year: 2021, Volume and Issue: 76, P. 103445 - 103445

Published: Oct. 13, 2021

The efficiency, flexibility, and resilience of building-integrated energy systems are challenged by unpredicted changes in operational environments due to climate change its consequences. On the other hand, rapid evolution artificial intelligence (AI) machine learning (ML) has equipped buildings with an ability learn. A lot research been dedicated specific applications for phases a building's life-cycle. reviews commonly take specific, technological perspective without vision integration smart technologies at level whole system. Especially, there is lack discussion on roles autonomous AI agents training boosting process complex abruptly changing environments. This review article discusses system-level presents overview that make independent decisions building management. We conclude buildings’ adaptability can be enhanced system through AI-initiated processes using digital twins as greatest potential efficiency improvement achieved integrating solutions timescales HVAC control electricity market participation.

Language: Английский

---

Reinforcement learning in sustainable energy and electric systems: a survey

Annual Reviews in Control, Journal Year: 2020, Volume and Issue: 49, P. 145 - 163

Published: Jan. 1, 2020

Language: Английский

---

LightLearn: An adaptive and occupant centered controller for lighting based on reinforcement learning

Building and Environment, Journal Year: 2018, Volume and Issue: 147, P. 397 - 414

Published: Oct. 20, 2018

Language: Английский

---

Transfer learning with deep neural networks for model predictive control of HVAC and natural ventilation in smart buildings

Journal of Cleaner Production, Journal Year: 2019, Volume and Issue: 254, P. 119866 - 119866

Published: Dec. 28, 2019

Language: Английский

---