Computer based automation and control systems are becoming increasingly important in smart sustainable buildings,often referred to as automated buildings(ABs),in order to automatically control,optimize and supervise a...Computer based automation and control systems are becoming increasingly important in smart sustainable buildings,often referred to as automated buildings(ABs),in order to automatically control,optimize and supervise a wide range of building performance applications over a network while minimizing energy consumption and associated green house gas emission.This technology generally refers to building automation and control systems(BACS)architecture.Instead of costly and time-consuming experiments,this paper focuses on development and design of a distributed dynamic simulation environment with the capability to represent BACS architecture in simulation by run-time coupling two or more different software tools over a network.This involves using distributed dynamic simulations as means to analyze the performance and enhance networked real-time control systems in ABs and improve the functions of real BACS technology.The application and capability of this new dynamic simulation environment are demonstrated by an experimental design,in this paper.展开更多
This work proposes an adaptive quantum approximate optimization-based model predictive control(MPC)strategy for energy management in buildings equipped with battery energy storage and renewable energy generation syste...This work proposes an adaptive quantum approximate optimization-based model predictive control(MPC)strategy for energy management in buildings equipped with battery energy storage and renewable energy generation systems.The learning-based parameter transfer scheme to realize adaptive quantum optimization leverages Bayesian optimization to predict initial quantum circuit parameters.When applied to the MPC problems formulated as quadratic unconstrained binary optimization problems,this approach computes optimal controls to minimize the net energy consumption levels in buildings and promotes decarbonization while reducing the computational efforts required for the quantum approximate optimization algorithm as the building energy system trajectory progresses.The energy efficiency and the decarbonization benefits of the proposed quantum optimization-based MPC strategy are demonstrated on buildings at the Cornell University campus.The proposed quantum computing-based technique to address MPC problems in buildings demonstrates energy-efficient and low-carbon building operation with a 6.8% improvement over deterministic MPC and presents opportunities for scaling to larger control problems with a significant reduction in utilized quantum computing resources.A reduction of 41.2% in carbon emissions is also achieved with the proposed control strategy facilitated by efficiently managing battery energy storage and renewable generation sources to promote a push toward carbonneutral building operations.展开更多
Buildings are a major energy consumer and carbon emitter,therefore it is important to improve building energy efficiency to achieve our sustainable development goal.Deep reinforcement learning(DRL),as an advanced buil...Buildings are a major energy consumer and carbon emitter,therefore it is important to improve building energy efficiency to achieve our sustainable development goal.Deep reinforcement learning(DRL),as an advanced building control method,demonstrates great potential for energy efficiency optimization and improved occupant comfort.However,the performance of DRL is highly sensitive to hyper-parameters,and selecting inappropriate hyper-parameters may lead to unstable learning or even failure.This study aims to investigate the design and application of DRL in building energy system control,with a specific focus on improving the performance of DRL controllers through hyper-parameter optimization(HPO)algorithms.It also aims to provide quantitative evaluation and adaptive validation of these optimized controllers.Two widely used algorithms,deep deterministic policy gradient(DDPG)and soft actor-critic(SAC),are used in the study and their performance is evaluated in different building environments based on the BOPTEST virtual testbed.One of the focuses of the study is to compare various HPO techniques,including tree-structured Parzen estimator(TPE),covariance matrix adaptation evolution strategy(CMA-ES),and combinatorial optimization methods,to determine the efficacy of different hyper-parameter optimization methods for DRL.The study enhances HPO efficiency through parallel computation and conducts a comprehensive quantitative assessment of the optimized DRL controllers,considering factors such as reduced energy consumption and improved comfort.The results show that the HPO algorithms significantly improve the performance of the DDPG and SAC controllers.A reduction of 56.94%and 68.74%in thermal discomfort is achieved,respectively.Additionally,the study demonstrates the applicability of the HPO-based approach for enhancing DRL controller performance across diverse building environments,providing valuable insights for the design and optimization of building DRL controllers.展开更多
The energy consumption of a teaching building can be effectively reduced by timetable optimization.However,in most studies that explore methods to reduce building energy consumption by course timetable optimization,se...The energy consumption of a teaching building can be effectively reduced by timetable optimization.However,in most studies that explore methods to reduce building energy consumption by course timetable optimization,self-study activities are not considered.In this study,an MATLAB-EnergyPlus joint simulation model was constructed based on the Building Controls Virtual Test Bed platform to reduce building energy consumption by optimizing the course schedule and opening strategy of self-study rooms in a holistic way.The following results were obtained by taking a university in Xi’an as an example:(1)The energy saving percentages obtained by timetabling optimization during the heating season examination week,heating season non-examination week,cooling season examination week,and cooling season non-examination week are 35%,29.4%,13.4%,and 13.4%,respectively.(2)Regarding the temporal arrangement,most courses are scheduled in the morning during the cooling season and afternoon during the heating season.Regarding the spatial arrangement,most courses are arranged in the central section of the middle floors of the building.(3)During the heating season,the additional building energy consumption incurred by the opening of self-study rooms decreases when duty heating temperature increases.展开更多
Reinforcement learning(RL)has shown significant success in sequential decision making in fields like autonomous vehicles,robotics,marketing and gaming industries.This success has attracted the attention to the RL cont...Reinforcement learning(RL)has shown significant success in sequential decision making in fields like autonomous vehicles,robotics,marketing and gaming industries.This success has attracted the attention to the RL control approach for building energy systems which are becoming complicated due to the need to optimize for multiple,potentially conflicting,goals like occupant comfort,energy use and grid interactivity.However,for real world applications,RL has several drawbacks like requiring large training data and time,and unstable control behavior during the early exploration process making it infeasible for an application directly to building control tasks.To address these issues,an imitation learning approach is utilized herein where the RL agents starts with a policy transferred from accepted rule based policies and heuristic policies.This approach is successful in reducing the training time,preventing the unstable early exploration behavior and improving upon an accepted rule-based policy-all of these make RL a more practical control approach for real world applications in the domain of building controls.展开更多
Occupant behavior in buildings has been considered the major source of uncertainty for assessing energy con-sumption and building performance.Modeling frameworks are usually built to accomplish a certain task,but the ...Occupant behavior in buildings has been considered the major source of uncertainty for assessing energy con-sumption and building performance.Modeling frameworks are usually built to accomplish a certain task,but the stochasticity of the occupant makes it difficult to apply that experience to a similar but distinct environment.For complex and dynamic environments,the development of smart devices and computing power makes intelligent control methods for occupant behaviors more viable.It is expected that they will make a substantial contribution to reducing global energy consumption.Among these control techniques,the reinforcement learning(RL)method seems distinctive and applicable.The success of the reinforcement learning method in many artificial intelligence applications has given an explicit indication of how this method might be used to model and adjust occupant behavior in building control.Fruitful algorithms complement each other and guarantee the quality of the opti-mization.However,the examination of occupant behavior based on reinforcement learning methodologies is not well established.The way that occupant interacts with the RL agent is still unclear.This study briefly reviews the empirical applications using reinforcement learning,how they have contributed to shaping the modeling paradigms and how they might suggest a future research direction.展开更多
This study proposes a refined methodology for controlling building heights in heritage areas.In order to protect the visual integrity of the heritage area,buildings should not obstruct the view from important site poi...This study proposes a refined methodology for controlling building heights in heritage areas.In order to protect the visual integrity of the heritage area,buildings should not obstruct the view from important site points and viewpoints to the periphery.By calculating the building height thresholds that buildings should not obscure the view from each viewpoint,the results of which are weighted and superimposed,and the values are extracted to each building unit as a refined building height control guideline.This study takes the Zhoukoudian area as a case study,applies the refined building height control criterion to the Zhoukoudian Site,and relies on this refined criterion to assess the visual integrity of the Zhoukoudian area,so as to realize the scientific planning and monitoring of the Zhoukoudian area.The refined building height control method can be applied to building height planning and visual landscape protection in large heritage areas.展开更多
This paper describes economical strategies to design blast resistant electrical substations and control buildings that are commonly used at industrial plants.Limited literature addressed design aspects for this class ...This paper describes economical strategies to design blast resistant electrical substations and control buildings that are commonly used at industrial plants.Limited literature addressed design aspects for this class of buildings.Furthermore,little guidelines are available in practice to regulate this type of steel construction.The first part of the paper overviews the architectural and structural layouts of electrical buildings.Blast resistance requirements for occupied control buildings are also discussed.Simplified multiple degrees of freedom(MDOF)dynamic model is also illustrated that can be utilized for analysis of the blast resistant buildings.The economical aspects and cost savings resulting in using mobile blast resistant buildings are discussed.The article also highlights the engineering challenges that are encountered in design of mobile electrical facilities.The transportation procedure and design requirements are briefly described.Guidelines are proposed to calculate the center of mass of the building combined with interior equipment.The proposed design concept for electrical and control buildings is cost effective and can be implemented in industry to reduce projects cost.展开更多
Increased electricity consumption combined with new forms of generation is testing the reliability of our grid infrastructure.This work describes a method to improve the reliability of the grid through large-scale adv...Increased electricity consumption combined with new forms of generation is testing the reliability of our grid infrastructure.This work describes a method to improve the reliability of the grid through large-scale advanced building control.This paper develops a bi-level distributed control framework to shift the load of 153 buildings to achieve a system-level objective of tracking a power reference signal.This bi-level control is based on the previously-developed ANPV-MPC,a predictive controller that uses a Bayesian neural network to generate an accurate control model and adapt to changing conditions over time.By shifting the building electricity demand to better match the available power,the grid system supplying the buildings is more reliable as evidenced by the analysis of node voltages across an IEEE 13-bus distribution system.The proposed bi-level control framework tracks the system-level power reference with enough accuracy to regulate node voltages across the IEEE 13-bus distribution system within ANSI limits of±5%.Additionally,the adaptive nature of ANPV-MPC allows each building across the system to adapt to changing conditions,further amplifying the system-level reliability.展开更多
The issue of weathertightness of the external building envelope in domestic scaled timber frames continues to be an issue in New Zealand, some ten years after the results of a major cladding survey into the durability...The issue of weathertightness of the external building envelope in domestic scaled timber frames continues to be an issue in New Zealand, some ten years after the results of a major cladding survey into the durability and weathertightness of the exterior cladding envelope carried out by the writer in the year 2000. The fallout from leaking buildings has estimated to have cost the country billions of dollars in lost production and expensive repair. The social impact on those caught up in the leaking home issue has been considerable; with often heart rending tales of stress and financial hardship. This paper will explore the initiatives taken by the building industry and the government since the issue became a major public concern. It will examine the influences, both positive and negative, that resulting legislation and changed building practices, brought in as a result of this crisis, have had on the sustainability and affordability of the domestic dwelling in New Zealand.展开更多
An experimental method is introduced in this paper to build the dynamics of AMSS (the active magnetic suspension system), which doesn’t depend on system’s physical parameters. The rotor can be reliably suspended und...An experimental method is introduced in this paper to build the dynamics of AMSS (the active magnetic suspension system), which doesn’t depend on system’s physical parameters. The rotor can be reliably suspended under the unit feedback control system designed with the primary dynamic model obtained. Online identification in frequency domain is processed to give the precise model. Comparisons show that the experimental method is much closer to the precise model than the theoretic method based on magnetic circuit law. So this experimental method is a good choice to build the primary dynamic model of AMSS.展开更多
Some building components are responsible for achieving more than one environmental function, these functions are usually of different requirements that can never be done by the same actions, and they are usually conne...Some building components are responsible for achieving more than one environmental function, these functions are usually of different requirements that can never be done by the same actions, and they are usually connected to changeable internal and external environment characteristics that vary among them. Minimizing the conflict of achieving the different environmental functions is an important challenge for all designers. Achieving a continuous thermal and optical comfort in an internal building space using the same window is an example of this challenge, as they have different requirements that may be sometimes contrary. It should be notable that there are a lot of recent technologies that may be used to find solutions for such a conflict. The Environmental Assessment Methods of Buildings appeared to set the principles of the optimum relation between buildings and their environment, they also could be used to encourage designers to reach the best environmental relations, and award them by main or additional assessment points. The research paper proposes to use the Environmental Assessment Methods of Buildings to assess the building ability of minimizing its environmental functions achievement conflict. This proposal depends on determining the inconsistency assessment items that depend on common building components to be achieved, and then determining the time periods that these items are achieved together within, to indicate the time periods without conflicting. Thus, the paper aims to raise the building environmental value in the assessment when the designer succeeds to minimize the expected conflict of the building environmental functions.展开更多
Management of the electrical grid is becoming more complex due to the increased penetration of alternative energy generation technologies and a broadening diversity of electric loads.This complexity creates challenges...Management of the electrical grid is becoming more complex due to the increased penetration of alternative energy generation technologies and a broadening diversity of electric loads.This complexity creates challenges in balancing demand and generation that can increase the potential for grid instabilities.One effective way to address this issue is to leverage previously unexploited demand flexibility through advanced control strategies.In this work,we propose an advanced control method,called adaptive neural parameter-varying model predictive control(ANPV-MPC),to control the temperature and energy consumption of a building via its Heating,Ventilation,and Air Conditioning system.ANPV-MPC combines key ideas in varying parameter-control,adaptive control,and online learning strategies to bridge the gap between computationally efficient linear model predictive control and more accurate nonlinear model predictive control.The novelty in ANPV-MPC is the use of a physics-inspired Bayesian neural network to estimate the coefficients of the parameter-varying linear control model.The Bayesian neural network additionally provides uncertainty estimates,triggering online training to capture evolving building system conditions.We show that ANPV-MPC can approximate the building system dynamics with a 28.39%higher accuracy than traditional linear model predictive control,resulting in 36.23%better control performance without increasing complexity of the optimal control problem.ANPV-MPC also adapts in real time to previously unseen conditions using online learning,further improving its performance.展开更多
Deep Reinforcement Learning(DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers(RBCs),but it still lacks scalability and generalisatio...Deep Reinforcement Learning(DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers(RBCs),but it still lacks scalability and generalisation due to the necessity of using tailored models for the training process.Transfer Learning(TL)is a potential solution to address this limitation.However,existing TL applications in building control have been mostly tested among buildings with similar features,not addressing the need to scale up advanced control in real-world scenarios with diverse energy systems.This paper assesses the performance of an online heterogeneous TL strategy,comparing it with RBC and offline and online DRL controllers in a simulation setup using EnergyPlus and Python.The study tests the transfer in both transductive and inductive settings of a DRL policy designed to manage a chiller coupled with a Thermal Energy Storage(TES).The control policy is pre-trained on a source building and transferred to various target buildings characterised by an integrated energy system including photovoltaic and battery energy storage systems,different building envelope features,occupancy schedule and boundary conditions(e.g.,weather and price signal).The TL approach incorporates model slicing,imitation learning and fine-tuning to handle diverse state spaces and reward functions between source and target buildings.Results show that the proposed methodology leads to a reduction of 10% in electricity cost and between 10% and 40% in the mean value of the daily average temperature violation rate compared to RBC and online DRL controllers.Moreover,online TL maximises self-sufficiency and self-consumption by 9% and 11% with respect to RBC.Conversely,online TL achieves worse performance compared to offline DRL in either transductive or inductive settings.However,offline Deep Reinforcement Learning(DRL)agents should be trained at least for 15 episodes to reach the same level of performance as the online TL.Therefore,the proposed online TL methodology is effective,completely model-free and it can be directly implemented in real buildings with satisfying performance.展开更多
Reinforcement learning(RL)has proven effective for AI-based building energy management.However,there is a lack of flexible framework to implement RL across various control problems in building energy management.To add...Reinforcement learning(RL)has proven effective for AI-based building energy management.However,there is a lack of flexible framework to implement RL across various control problems in building energy management.To address this gap,we propose BuildingGym,an open-source tool designed as a research-friendly and flexible framework for training RL control strategies for common challenges in building energy management.BuildingGym integrates EnergyPlus as its core simulator,making it suitable for both system-level and room-level control.Additionally,BuildingGym is able to accept external signals as control inputs instead of taking the building as a stand-alone entity.This feature makes BuildingGym applicable for more flexible environments,e.g.smart grid and EVs community.The tool provides several built-in RL algorithms for control strategy training,simplifying the process for building managers to obtain optimal control strategies.Users can achieve this by following a few straightforward steps to configure BuildingGym for optimization control for common problems in the building energy management field.Moreover,AI specialists can easily implement and test state-of-the-art control algorithms within the platform.BuildingGym bridges the gap between building managers and AI specialists by allowing for the easy configuration and replacement of RL algorithms,simulators,and control environments or problems.With BuildingGym,we efficiently set up training tasks for cooling load management,targeting both constant and dynamic cooling load management.The built-in algorithms demonstrated strong performance across both tasks,highlighting the effectiveness of BuildingGym in optimizing cooling strategies.展开更多
Occupant-centric building simulation models rely on two key factors:our understanding of the underlying physics that govern thermal dynamics,and realistic modeling of occupancy patterns and energy use within the zone ...Occupant-centric building simulation models rely on two key factors:our understanding of the underlying physics that govern thermal dynamics,and realistic modeling of occupancy patterns and energy use within the zone of interest.While current physics-oriented building simulation models predict thermal dynamics accurately,a systematic and scalable way to generate occupancy and energy use patterns remains an open challenge despite the large amount of data collected from building sensors across academic and industry efforts.In this paper,we leverage deep generative networks capable of learning from real building data for generating realistic occupant-centric scenarios to inform building simulations.Our ultimate goal is to assess building performance over a wide range of generated scenarios,which is currently done either by taking a small set of“nominal scenarios”or by handcrafting specific scenarios,both of which restrict the quality of building performance assessment to a few biased use-cases.For the purpose of generating scenarios automatically,we employ a recently proposed architecture called RAFT-VG(regularized adversarially fine-tuned VAE-GAN)that combines the benefits of variational autoencoders(VAEs)and generative adversarial networks(GANs),and demonstrate its capacity for synthesizing a variety of signals including occupancy patterns,internal heat loads,and ambient conditions.A key feature of this neural architecture is that the generative process depends solely on a conditional decoder network.Distilling the deep RAFT-VG model to a simpler decoder for inference allows us to propose a general framework for integrating the generative model directly in Modelica.The closed-loop building performance with various generated scenarios,along with the Modelica integration,is demonstrated via simulation use-cases using the Modelica BESTEST repository.展开更多
Mechanical ventilation is an effective measure to control indoor long-range airborne transmission of COVID-19,but it often leads to substantial energy expenditure.This study introduces a novel exposure-based smart ven...Mechanical ventilation is an effective measure to control indoor long-range airborne transmission of COVID-19,but it often leads to substantial energy expenditure.This study introduces a novel exposure-based smart ventilation and occupancy control strategy to reduce infection risk and save energy in school environments that are typically characterized by fixed occupants and long exposure time.This exposure-based approach allows the quanta concentration to vary over time rather than keeping it constantly below certain thresholds.This enables us to:(1)adjust ventilation and occupant schedule to facilitate passive cooling/heating potential in response to outdoor weather conditions;(2)consider the interaction between ventilation and occupant schedule to maximize their benefits in reducing infection risk and energy consumption.Taking a typical classroom as a base case,ventilation and occupant schedule are optimized individually and jointly through Genetic Algorithm,to control infection risk,minimize energy consumption,maintain thermal comfort,and promise sufficient schooling time.Our results show that the most energy-efficient strategy is the concurrent optimization of both occupant schedule and ventilation,achieving an energy reduction of up to~60%compared to traditional constant ventilation methods.Solely optimizing occupant schedule is the least energy-efficient strategy,yielding an energy reduction ratio(over base case)of only half of the most efficient strategy.Our study reveals the possibility of optimizing occupant schedule and ventilation to balance building energy consumption and transmission control.The viability of these control strategies has been proven across various climate zones and seasons in China,highlighting their broad applicability.展开更多
Machine learning control(MLC)is a highly flexible and adaptable method that enables the design,modeling,tuning,and maintenance of building controllers to be more accurate,automated,flexible,and adaptable.The research ...Machine learning control(MLC)is a highly flexible and adaptable method that enables the design,modeling,tuning,and maintenance of building controllers to be more accurate,automated,flexible,and adaptable.The research topic of MLC in building energy systems is developing rapidly,but to our knowledge,no review has been published that specifically and systematically focuses on MLC for building energy systems.This paper provides a systematic review of MLC in building energy systems.We review technical papers in two major categories of applications of machine learning in building control:(1)building system and component modeling for control,and(2)control process learning.We identify MLC topics that have been well-studied and those that need further research in the field of building operation control.We also identify the gaps between the present and future application of MLC and predict future trends and opportunities.展开更多
Model predictive control(MPC)is an advanced control technique.It has been deployed to harness the energy flexibility of a building.MPC requires a dynamic model of the building to achieve such an objective.However,deve...Model predictive control(MPC)is an advanced control technique.It has been deployed to harness the energy flexibility of a building.MPC requires a dynamic model of the building to achieve such an objective.However,developing a suitable predictive model is the main challenge in MPC implementation forflexibility activation.This studyfocuses on the application of key performance indicators(KPls)to evaluate the suitability of MPC models via feature selection.To this end,multiple models were developed for two houses.A feature selection method was developed to select an appropriate feature space to train the models.These predictive models were then quantified based on one-step ahead prediction error(OSPE),a standard KPI used in multiple studies,and a less-often KPl:multi-step ahead prediction error(MSPE).An MPC workflow was designed where different models can serve as the predictive model.Findings showed that MSPE better demonstrates the performance of predictive models used for flexibility activation.Results revealed that up to 57% of the flexibility potential and 48% of energy use reduction are not exploited if MSPE is not minimized while developing a predictive model.展开更多
文摘Computer based automation and control systems are becoming increasingly important in smart sustainable buildings,often referred to as automated buildings(ABs),in order to automatically control,optimize and supervise a wide range of building performance applications over a network while minimizing energy consumption and associated green house gas emission.This technology generally refers to building automation and control systems(BACS)architecture.Instead of costly and time-consuming experiments,this paper focuses on development and design of a distributed dynamic simulation environment with the capability to represent BACS architecture in simulation by run-time coupling two or more different software tools over a network.This involves using distributed dynamic simulations as means to analyze the performance and enhance networked real-time control systems in ABs and improve the functions of real BACS technology.The application and capability of this new dynamic simulation environment are demonstrated by an experimental design,in this paper.
文摘This work proposes an adaptive quantum approximate optimization-based model predictive control(MPC)strategy for energy management in buildings equipped with battery energy storage and renewable energy generation systems.The learning-based parameter transfer scheme to realize adaptive quantum optimization leverages Bayesian optimization to predict initial quantum circuit parameters.When applied to the MPC problems formulated as quadratic unconstrained binary optimization problems,this approach computes optimal controls to minimize the net energy consumption levels in buildings and promotes decarbonization while reducing the computational efforts required for the quantum approximate optimization algorithm as the building energy system trajectory progresses.The energy efficiency and the decarbonization benefits of the proposed quantum optimization-based MPC strategy are demonstrated on buildings at the Cornell University campus.The proposed quantum computing-based technique to address MPC problems in buildings demonstrates energy-efficient and low-carbon building operation with a 6.8% improvement over deterministic MPC and presents opportunities for scaling to larger control problems with a significant reduction in utilized quantum computing resources.A reduction of 41.2% in carbon emissions is also achieved with the proposed control strategy facilitated by efficiently managing battery energy storage and renewable generation sources to promote a push toward carbonneutral building operations.
基金supported by the National Natural Science Foundation of China(No.72371072)Jiangsu Association for Science&Technology Youth Science&Technology Talents Lifting Project(No.JSTJ-2023-JS001).
文摘Buildings are a major energy consumer and carbon emitter,therefore it is important to improve building energy efficiency to achieve our sustainable development goal.Deep reinforcement learning(DRL),as an advanced building control method,demonstrates great potential for energy efficiency optimization and improved occupant comfort.However,the performance of DRL is highly sensitive to hyper-parameters,and selecting inappropriate hyper-parameters may lead to unstable learning or even failure.This study aims to investigate the design and application of DRL in building energy system control,with a specific focus on improving the performance of DRL controllers through hyper-parameter optimization(HPO)algorithms.It also aims to provide quantitative evaluation and adaptive validation of these optimized controllers.Two widely used algorithms,deep deterministic policy gradient(DDPG)and soft actor-critic(SAC),are used in the study and their performance is evaluated in different building environments based on the BOPTEST virtual testbed.One of the focuses of the study is to compare various HPO techniques,including tree-structured Parzen estimator(TPE),covariance matrix adaptation evolution strategy(CMA-ES),and combinatorial optimization methods,to determine the efficacy of different hyper-parameter optimization methods for DRL.The study enhances HPO efficiency through parallel computation and conducts a comprehensive quantitative assessment of the optimized DRL controllers,considering factors such as reduced energy consumption and improved comfort.The results show that the HPO algorithms significantly improve the performance of the DDPG and SAC controllers.A reduction of 56.94%and 68.74%in thermal discomfort is achieved,respectively.Additionally,the study demonstrates the applicability of the HPO-based approach for enhancing DRL controller performance across diverse building environments,providing valuable insights for the design and optimization of building DRL controllers.
基金supported by the National Natural Science Foundation of China (52008328)National Key Research and Development Project (2018YFD1100202)+1 种基金the Science and Technology Department of Shaanxi Province (2020SF-393,2018ZDCXL-SF-03-04)the State Key Laboratory of Green Building in Western China (LSZZ202009).
文摘The energy consumption of a teaching building can be effectively reduced by timetable optimization.However,in most studies that explore methods to reduce building energy consumption by course timetable optimization,self-study activities are not considered.In this study,an MATLAB-EnergyPlus joint simulation model was constructed based on the Building Controls Virtual Test Bed platform to reduce building energy consumption by optimizing the course schedule and opening strategy of self-study rooms in a holistic way.The following results were obtained by taking a university in Xi’an as an example:(1)The energy saving percentages obtained by timetabling optimization during the heating season examination week,heating season non-examination week,cooling season examination week,and cooling season non-examination week are 35%,29.4%,13.4%,and 13.4%,respectively.(2)Regarding the temporal arrangement,most courses are scheduled in the morning during the cooling season and afternoon during the heating season.Regarding the spatial arrangement,most courses are arranged in the central section of the middle floors of the building.(3)During the heating season,the additional building energy consumption incurred by the opening of self-study rooms decreases when duty heating temperature increases.
基金This work was authored in part by the National Renewable Energy Laboratory,United States,operated by Alliance for Sustainable Energy,LLC,for the U.S.Department of Energy(DOE)under Contract No.DE-AC36-08GO28308.
文摘Reinforcement learning(RL)has shown significant success in sequential decision making in fields like autonomous vehicles,robotics,marketing and gaming industries.This success has attracted the attention to the RL control approach for building energy systems which are becoming complicated due to the need to optimize for multiple,potentially conflicting,goals like occupant comfort,energy use and grid interactivity.However,for real world applications,RL has several drawbacks like requiring large training data and time,and unstable control behavior during the early exploration process making it infeasible for an application directly to building control tasks.To address these issues,an imitation learning approach is utilized herein where the RL agents starts with a policy transferred from accepted rule based policies and heuristic policies.This approach is successful in reducing the training time,preventing the unstable early exploration behavior and improving upon an accepted rule-based policy-all of these make RL a more practical control approach for real world applications in the domain of building controls.
基金The authors are thankful for the financial support from IMMA project of research network(391836)Dalarna University,Sweden and Inter-national science and technology cooperation center in Hebei Province(20594501D),China.
文摘Occupant behavior in buildings has been considered the major source of uncertainty for assessing energy con-sumption and building performance.Modeling frameworks are usually built to accomplish a certain task,but the stochasticity of the occupant makes it difficult to apply that experience to a similar but distinct environment.For complex and dynamic environments,the development of smart devices and computing power makes intelligent control methods for occupant behaviors more viable.It is expected that they will make a substantial contribution to reducing global energy consumption.Among these control techniques,the reinforcement learning(RL)method seems distinctive and applicable.The success of the reinforcement learning method in many artificial intelligence applications has given an explicit indication of how this method might be used to model and adjust occupant behavior in building control.Fruitful algorithms complement each other and guarantee the quality of the opti-mization.However,the examination of occupant behavior based on reinforcement learning methodologies is not well established.The way that occupant interacts with the RL agent is still unclear.This study briefly reviews the empirical applications using reinforcement learning,how they have contributed to shaping the modeling paradigms and how they might suggest a future research direction.
文摘This study proposes a refined methodology for controlling building heights in heritage areas.In order to protect the visual integrity of the heritage area,buildings should not obstruct the view from important site points and viewpoints to the periphery.By calculating the building height thresholds that buildings should not obscure the view from each viewpoint,the results of which are weighted and superimposed,and the values are extracted to each building unit as a refined building height control guideline.This study takes the Zhoukoudian area as a case study,applies the refined building height control criterion to the Zhoukoudian Site,and relies on this refined criterion to assess the visual integrity of the Zhoukoudian area,so as to realize the scientific planning and monitoring of the Zhoukoudian area.The refined building height control method can be applied to building height planning and visual landscape protection in large heritage areas.
文摘This paper describes economical strategies to design blast resistant electrical substations and control buildings that are commonly used at industrial plants.Limited literature addressed design aspects for this class of buildings.Furthermore,little guidelines are available in practice to regulate this type of steel construction.The first part of the paper overviews the architectural and structural layouts of electrical buildings.Blast resistance requirements for occupied control buildings are also discussed.Simplified multiple degrees of freedom(MDOF)dynamic model is also illustrated that can be utilized for analysis of the blast resistant buildings.The economical aspects and cost savings resulting in using mobile blast resistant buildings are discussed.The article also highlights the engineering challenges that are encountered in design of mobile electrical facilities.The transportation procedure and design requirements are briefly described.Guidelines are proposed to calculate the center of mass of the building combined with interior equipment.The proposed design concept for electrical and control buildings is cost effective and can be implemented in industry to reduce projects cost.
文摘Increased electricity consumption combined with new forms of generation is testing the reliability of our grid infrastructure.This work describes a method to improve the reliability of the grid through large-scale advanced building control.This paper develops a bi-level distributed control framework to shift the load of 153 buildings to achieve a system-level objective of tracking a power reference signal.This bi-level control is based on the previously-developed ANPV-MPC,a predictive controller that uses a Bayesian neural network to generate an accurate control model and adapt to changing conditions over time.By shifting the building electricity demand to better match the available power,the grid system supplying the buildings is more reliable as evidenced by the analysis of node voltages across an IEEE 13-bus distribution system.The proposed bi-level control framework tracks the system-level power reference with enough accuracy to regulate node voltages across the IEEE 13-bus distribution system within ANSI limits of±5%.Additionally,the adaptive nature of ANPV-MPC allows each building across the system to adapt to changing conditions,further amplifying the system-level reliability.
文摘The issue of weathertightness of the external building envelope in domestic scaled timber frames continues to be an issue in New Zealand, some ten years after the results of a major cladding survey into the durability and weathertightness of the exterior cladding envelope carried out by the writer in the year 2000. The fallout from leaking buildings has estimated to have cost the country billions of dollars in lost production and expensive repair. The social impact on those caught up in the leaking home issue has been considerable; with often heart rending tales of stress and financial hardship. This paper will explore the initiatives taken by the building industry and the government since the issue became a major public concern. It will examine the influences, both positive and negative, that resulting legislation and changed building practices, brought in as a result of this crisis, have had on the sustainability and affordability of the domestic dwelling in New Zealand.
基金Supported by the National Nature Foundation of China (No.59975073)
文摘An experimental method is introduced in this paper to build the dynamics of AMSS (the active magnetic suspension system), which doesn’t depend on system’s physical parameters. The rotor can be reliably suspended under the unit feedback control system designed with the primary dynamic model obtained. Online identification in frequency domain is processed to give the precise model. Comparisons show that the experimental method is much closer to the precise model than the theoretic method based on magnetic circuit law. So this experimental method is a good choice to build the primary dynamic model of AMSS.
文摘Some building components are responsible for achieving more than one environmental function, these functions are usually of different requirements that can never be done by the same actions, and they are usually connected to changeable internal and external environment characteristics that vary among them. Minimizing the conflict of achieving the different environmental functions is an important challenge for all designers. Achieving a continuous thermal and optical comfort in an internal building space using the same window is an example of this challenge, as they have different requirements that may be sometimes contrary. It should be notable that there are a lot of recent technologies that may be used to find solutions for such a conflict. The Environmental Assessment Methods of Buildings appeared to set the principles of the optimum relation between buildings and their environment, they also could be used to encourage designers to reach the best environmental relations, and award them by main or additional assessment points. The research paper proposes to use the Environmental Assessment Methods of Buildings to assess the building ability of minimizing its environmental functions achievement conflict. This proposal depends on determining the inconsistency assessment items that depend on common building components to be achieved, and then determining the time periods that these items are achieved together within, to indicate the time periods without conflicting. Thus, the paper aims to raise the building environmental value in the assessment when the designer succeeds to minimize the expected conflict of the building environmental functions.
基金supported by the Laboratory Directed Research and Development(LDRD)Program at NREL.
文摘Management of the electrical grid is becoming more complex due to the increased penetration of alternative energy generation technologies and a broadening diversity of electric loads.This complexity creates challenges in balancing demand and generation that can increase the potential for grid instabilities.One effective way to address this issue is to leverage previously unexploited demand flexibility through advanced control strategies.In this work,we propose an advanced control method,called adaptive neural parameter-varying model predictive control(ANPV-MPC),to control the temperature and energy consumption of a building via its Heating,Ventilation,and Air Conditioning system.ANPV-MPC combines key ideas in varying parameter-control,adaptive control,and online learning strategies to bridge the gap between computationally efficient linear model predictive control and more accurate nonlinear model predictive control.The novelty in ANPV-MPC is the use of a physics-inspired Bayesian neural network to estimate the coefficients of the parameter-varying linear control model.The Bayesian neural network additionally provides uncertainty estimates,triggering online training to capture evolving building system conditions.We show that ANPV-MPC can approximate the building system dynamics with a 28.39%higher accuracy than traditional linear model predictive control,resulting in 36.23%better control performance without increasing complexity of the optimal control problem.ANPV-MPC also adapts in real time to previously unseen conditions using online learning,further improving its performance.
基金funded by the project NODES which has received funding from the MUR-M4C21.5 of PNRR funded by the European Union-NextGenerationEU(Grant agreement no.ECS00000036).
文摘Deep Reinforcement Learning(DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers(RBCs),but it still lacks scalability and generalisation due to the necessity of using tailored models for the training process.Transfer Learning(TL)is a potential solution to address this limitation.However,existing TL applications in building control have been mostly tested among buildings with similar features,not addressing the need to scale up advanced control in real-world scenarios with diverse energy systems.This paper assesses the performance of an online heterogeneous TL strategy,comparing it with RBC and offline and online DRL controllers in a simulation setup using EnergyPlus and Python.The study tests the transfer in both transductive and inductive settings of a DRL policy designed to manage a chiller coupled with a Thermal Energy Storage(TES).The control policy is pre-trained on a source building and transferred to various target buildings characterised by an integrated energy system including photovoltaic and battery energy storage systems,different building envelope features,occupancy schedule and boundary conditions(e.g.,weather and price signal).The TL approach incorporates model slicing,imitation learning and fine-tuning to handle diverse state spaces and reward functions between source and target buildings.Results show that the proposed methodology leads to a reduction of 10% in electricity cost and between 10% and 40% in the mean value of the daily average temperature violation rate compared to RBC and online DRL controllers.Moreover,online TL maximises self-sufficiency and self-consumption by 9% and 11% with respect to RBC.Conversely,online TL achieves worse performance compared to offline DRL in either transductive or inductive settings.However,offline Deep Reinforcement Learning(DRL)agents should be trained at least for 15 episodes to reach the same level of performance as the online TL.Therefore,the proposed online TL methodology is effective,completely model-free and it can be directly implemented in real buildings with satisfying performance.
基金supported by AI Singapore under its project of Development of NetZero BEMS through AI-based HVAC System Control(AISG2-TC-2023-008-SGKR).
文摘Reinforcement learning(RL)has proven effective for AI-based building energy management.However,there is a lack of flexible framework to implement RL across various control problems in building energy management.To address this gap,we propose BuildingGym,an open-source tool designed as a research-friendly and flexible framework for training RL control strategies for common challenges in building energy management.BuildingGym integrates EnergyPlus as its core simulator,making it suitable for both system-level and room-level control.Additionally,BuildingGym is able to accept external signals as control inputs instead of taking the building as a stand-alone entity.This feature makes BuildingGym applicable for more flexible environments,e.g.smart grid and EVs community.The tool provides several built-in RL algorithms for control strategy training,simplifying the process for building managers to obtain optimal control strategies.Users can achieve this by following a few straightforward steps to configure BuildingGym for optimization control for common problems in the building energy management field.Moreover,AI specialists can easily implement and test state-of-the-art control algorithms within the platform.BuildingGym bridges the gap between building managers and AI specialists by allowing for the easy configuration and replacement of RL algorithms,simulators,and control environments or problems.With BuildingGym,we efficiently set up training tasks for cooling load management,targeting both constant and dynamic cooling load management.The built-in algorithms demonstrated strong performance across both tasks,highlighting the effectiveness of BuildingGym in optimizing cooling strategies.
文摘Occupant-centric building simulation models rely on two key factors:our understanding of the underlying physics that govern thermal dynamics,and realistic modeling of occupancy patterns and energy use within the zone of interest.While current physics-oriented building simulation models predict thermal dynamics accurately,a systematic and scalable way to generate occupancy and energy use patterns remains an open challenge despite the large amount of data collected from building sensors across academic and industry efforts.In this paper,we leverage deep generative networks capable of learning from real building data for generating realistic occupant-centric scenarios to inform building simulations.Our ultimate goal is to assess building performance over a wide range of generated scenarios,which is currently done either by taking a small set of“nominal scenarios”or by handcrafting specific scenarios,both of which restrict the quality of building performance assessment to a few biased use-cases.For the purpose of generating scenarios automatically,we employ a recently proposed architecture called RAFT-VG(regularized adversarially fine-tuned VAE-GAN)that combines the benefits of variational autoencoders(VAEs)and generative adversarial networks(GANs),and demonstrate its capacity for synthesizing a variety of signals including occupancy patterns,internal heat loads,and ambient conditions.A key feature of this neural architecture is that the generative process depends solely on a conditional decoder network.Distilling the deep RAFT-VG model to a simpler decoder for inference allows us to propose a general framework for integrating the generative model directly in Modelica.The closed-loop building performance with various generated scenarios,along with the Modelica integration,is demonstrated via simulation use-cases using the Modelica BESTEST repository.
基金support from China Scholarship Council(CSC)for pursuing her PhD at the University of Reading,UK.
文摘Mechanical ventilation is an effective measure to control indoor long-range airborne transmission of COVID-19,but it often leads to substantial energy expenditure.This study introduces a novel exposure-based smart ventilation and occupancy control strategy to reduce infection risk and save energy in school environments that are typically characterized by fixed occupants and long exposure time.This exposure-based approach allows the quanta concentration to vary over time rather than keeping it constantly below certain thresholds.This enables us to:(1)adjust ventilation and occupant schedule to facilitate passive cooling/heating potential in response to outdoor weather conditions;(2)consider the interaction between ventilation and occupant schedule to maximize their benefits in reducing infection risk and energy consumption.Taking a typical classroom as a base case,ventilation and occupant schedule are optimized individually and jointly through Genetic Algorithm,to control infection risk,minimize energy consumption,maintain thermal comfort,and promise sufficient schooling time.Our results show that the most energy-efficient strategy is the concurrent optimization of both occupant schedule and ventilation,achieving an energy reduction of up to~60%compared to traditional constant ventilation methods.Solely optimizing occupant schedule is the least energy-efficient strategy,yielding an energy reduction ratio(over base case)of only half of the most efficient strategy.Our study reveals the possibility of optimizing occupant schedule and ventilation to balance building energy consumption and transmission control.The viability of these control strategies has been proven across various climate zones and seasons in China,highlighting their broad applicability.
文摘Machine learning control(MLC)is a highly flexible and adaptable method that enables the design,modeling,tuning,and maintenance of building controllers to be more accurate,automated,flexible,and adaptable.The research topic of MLC in building energy systems is developing rapidly,but to our knowledge,no review has been published that specifically and systematically focuses on MLC for building energy systems.This paper provides a systematic review of MLC in building energy systems.We review technical papers in two major categories of applications of machine learning in building control:(1)building system and component modeling for control,and(2)control process learning.We identify MLC topics that have been well-studied and those that need further research in the field of building operation control.We also identify the gaps between the present and future application of MLC and predict future trends and opportunities.
基金funded by the Research Foundation Flanders(FWO),application number GOD2519Nby KU Leuven,grant C24/18/040.
文摘Model predictive control(MPC)is an advanced control technique.It has been deployed to harness the energy flexibility of a building.MPC requires a dynamic model of the building to achieve such an objective.However,developing a suitable predictive model is the main challenge in MPC implementation forflexibility activation.This studyfocuses on the application of key performance indicators(KPls)to evaluate the suitability of MPC models via feature selection.To this end,multiple models were developed for two houses.A feature selection method was developed to select an appropriate feature space to train the models.These predictive models were then quantified based on one-step ahead prediction error(OSPE),a standard KPI used in multiple studies,and a less-often KPl:multi-step ahead prediction error(MSPE).An MPC workflow was designed where different models can serve as the predictive model.Findings showed that MSPE better demonstrates the performance of predictive models used for flexibility activation.Results revealed that up to 57% of the flexibility potential and 48% of energy use reduction are not exploited if MSPE is not minimized while developing a predictive model.
