Model-free control method based on reinforcement learning for building cooling water systems: Validation by measured data-based simulation

Qiu, Shunian; Li, Zhenhai; Li, Jiajie; Long, Shengping; Li, Xiaoping

doi:10.1016/j.enbuild.2020.110055

Cited by 55 publications

(29 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead of a priori knowledge, the agent's actions are based on his or her own experience gained during the game. Research has been done in the field of building control to apply RL to model-based control [15]. In this research, the environment is the model for the airport building described previously.…”

Section: Building System Control Of the Reinforcement Learning (Rl) P...mentioning

confidence: 99%

See 1 more Smart Citation

Energy Saving by Reinforcement Learning for Multi-Chillers of HVAC Systems

Hussein¹,

Ateeq²,

Homod³

2022

Proceedings of 2nd International Multi-Disciplinary Conference Theme: Integrated Sciences and Technologies, IMDC-IST 2021, 7-9

View full text Add to dashboard Cite

This paper presents a method for controlling and operating a multi-chillers system: (1) Model-based control approach was used by MATLAB/SIMULINK to model a building containing two non-identical chillers depending on thermal loads. (2) ON/OFF all chillers alternately using the model reinforcement learning controller (RL-control) to select the appropriate chiller for the building conditioning process. The results were in terms of energy efficiency and performance of the enhanced learning control for the chiller, and a control unit signal (PID) was applied to make a comparison with the signals of energy, power, and temperatures. After comparison, it was found that the energy saving through the proposed controller is 45% of the traditional (PID) strategy, where can the proposed strategy control for the chiller appropriate for the building's conditioning process.

show abstract

Section: Building System Control Of the Reinforcement Learning (Rl) P...mentioning

confidence: 99%

“…(2020) a model-free optimal control method depends on RL is proposed to control building refrigerating water, [15]. Homod et al (2021) the HVAC system established was adopted as the system performance model; it provided dynamic system modelling, simulation, and monitoring cooling system performance.…”

Section: Introductionmentioning

confidence: 99%

Energy Saving by Reinforcement Learning for Multi-Chillers of HVAC Systems

Hussein¹,

Ateeq²,

Homod³

2022

Proceedings of 2nd International Multi-Disciplinary Conference Theme: Integrated Sciences and Technologies, IMDC-IST 2021, 7-9

View full text Add to dashboard Cite

show abstract

“…31 For the model of optimal control, adaptive modelling and model-free methods have been investigated for optimal control of the chiller plant. 32 Due to the proliferation of BMS, the size of building operation data increases exponentially, and data mining techniques can be used to supervise and improve the chiller plant operation. 33 With the efficient optimisation strategy and the adaptive or self-learning modelling technique, a dynamic optimal performance of the chiller plant operation can be achieved through an intelligent operation platform.…”

Section: Intelligent Operation Of Chiller Plantmentioning

confidence: 99%

Technology pathway of efficient and climate-friendly cooling in buildings: Towards carbon neutrality

Wang

Yu²,

Cao

2021

Indoor and Built Environment

View full text Add to dashboard Cite

The term 'carbon neutrality' was originally introduced in the United Nations Framework Convention on Climate Change -UNFCCC COP21 Paris Climate Accord, December 2015. 1 This has become a household buzzword in recent years. Carbon neutrality is a balance between carbon source reduction and carbon absorption, i.e. reducing carbon emission sources while also having an important consideration on carbon elimination and absorption. On 22 September 2020, the President Xi of China stated that 2 'Achieving carbon peaking and carbon neutrality is a profound economic and social revolution, and it should be incorporated into the overall planning of ecological civilization construction, so as to achieve carbon peaking by 2030 and carbon neutrality by 2060'.China's pledge to bring forward its strengthened national plan is a major boost to the progress of the upcoming UN climate congress, COP26 to be hosted by the UK in Glasgow in November 2021. China's carbon neutrality goal is feasible both technically and economically. 3,4 China, UK and the EU have set stronger commitment targets for reduction of carbon emission. China pledges: [4][5][6] China has committed to peak its CO2 emissions by 2030 and to make best efforts to peak earlier. By 2030, the country will increase its energy consumption of non-fossil fuel to 20%, and the country will install 800 to 1,000 gigawatts in non-fossil energy generating capacity, which is equivalent to the entire current US energy generation capacity. China has put a price on carbon by committing to build a national cap and trade system. This was launched in 2017 and will become the world's largest. ¥20 billion RMB ($3.1 billion USD) has been pledged to the South-South Climate Cooperation Fund to help developing countries to tackle the climate change.

show abstract

“…The first RL application in the energy and building research field dates back to 1998, in which [24] RL was used on an HVAC system serving the DOE (Department of Energy) at Idaho State University, to control the water temperature leaving boiler and the temperature in the two thermal zones served by the system. The applications of DRL for the control of HVAC systems have increased considerably, with the aim of regulating different parameters such as: supply water temperature setpoint [5,25,26], storage tank temperature setpoint [27][28][29] supply air-flow rate [30], supply air temperature [31], indoor temperature setpoint [29,[32][33][34], frequency of pumps and fans [35,36]. Zhang et al [5] applied a DRL control type Asynchronous Advantage Actor-Critic (A3C) in a water-based Radiant Heating System, in which the hot water pipes were integrated into window mullions with the objective of reducing the energy consumption of the system while ensuring the internal comfort of the occupants.…”

Section: Related Work Motivations and Novelty Of The Papermentioning

confidence: 99%

Online Implementation of a Soft Actor-Critic Agent to Enhance Indoor Temperature Control and Energy Efficiency in Buildings

et al. 2021

View full text Add to dashboard Cite

Recently, a growing interest has been observed in HVAC control systems based on Artificial Intelligence, to improve comfort conditions while avoiding unnecessary energy consumption. In this work, a model-free algorithm belonging to the Deep Reinforcement Learning (DRL) class, Soft Actor-Critic, was implemented to control the supply water temperature to radiant terminal units of a heating system serving an office building. The controller was trained online, and a preliminary sensitivity analysis on hyperparameters was performed to assess their influence on the agent performance. The DRL agent with the best performance was compared to a rule-based controller assumed as a baseline during a three-month heating season. The DRL controller outperformed the baseline after two weeks of deployment, with an overall performance improvement related to control of indoor temperature conditions. Moreover, the adaptability of the DRL agent was tested for various control scenarios, simulating changes of external weather conditions, indoor temperature setpoint, building envelope features and occupancy patterns. The agent dynamically deployed, despite a slight increase in energy consumption, led to an improvement of indoor temperature control, reducing the cumulative sum of temperature violations on average for all scenarios by 75% and 48% compared to the baseline and statically deployed agent respectively.

show abstract

Model-free control method based on reinforcement learning for building cooling water systems: Validation by measured data-based simulation

Cited by 55 publications

References 43 publications

Energy Saving by Reinforcement Learning for Multi-Chillers of HVAC Systems

Energy Saving by Reinforcement Learning for Multi-Chillers of HVAC Systems

Technology pathway of efficient and climate-friendly cooling in buildings: Towards carbon neutrality

Online Implementation of a Soft Actor-Critic Agent to Enhance Indoor Temperature Control and Energy Efficiency in Buildings

Contact Info

Product

Resources

About