•Reinforcement learning has been explored for building control applications.•We reviewed studies using reinforcement learning for building controls.•We surveyed algorithm, state, action, reward, and ...the environment for reinforcement learning controller.•Research trends, progress and gaps of this field have been identified.•Adoption of reinforcement learning based controls in real buildings still faces significant challenges.
Building controls are becoming more important and complicated due to the dynamic and stochastic energy demand, on-site intermittent energy supply, as well as energy storage, making it difficult for them to be optimized by conventional control techniques. Reinforcement Learning (RL), as an emerging control technique, has attracted growing research interest and demonstrated its potential to enhance building performance while addressing some limitations of other advanced control techniques, such as model predictive control. This study conducted a comprehensive review of existing studies that applied RL for building controls. It provided a detailed breakdown of the existing RL studies that use a specific variation of each major component of the Reinforcement Learning: algorithm, state, action, reward, and environment. We found RL for building controls is still in the research stage with limited applications (11%) in real buildings. Three significant barriers prevent the adoption of RL controllers in actual building controls: (1) the training process is time consuming and data demanding, (2) the control security and robustness need to be enhanced, and (3) the generalization capabilities of RL controllers need to be improved using approaches such as transfer learning. Future research may focus on developing RL controllers that could be used in real buildings, addressing current RL challenges, such as accelerating training and enhancing control robustness, as well as developing an open-source testbed and dataset for performance benchmarking of RL controllers.
•Building load prediction informs chiller plant and thermal storage optimization.•We used and compared 9 machine learning algorithms and 3 heuristic prediction methods.•XGBoost and LSTM are found to ...be the best shallow and deep learning algorithm.•LSTM is better for short term prediction, while XGBoost for long term prediction.•It is better to train the model with uncertain rather than accurate weather data.
Building thermal load prediction informs the optimization of cooling plant and thermal energy storage. Physics-based prediction models of building thermal load are constrained by the model and input complexity. In this study, we developed 12 data-driven models (7 shallow learning, 2 deep learning, and 3 heuristic methods) to predict building thermal load and compared shallow machine learning and deep learning. The 12 prediction models were compared with the measured cooling demand. It was found XGBoost (Extreme Gradient Boost) and LSTM (Long Short Term Memory) provided the most accurate load prediction in the shallow and deep learning category, and both outperformed the best baseline model, which uses the previous day’s data for prediction. Then, we discussed how the prediction horizon and input uncertainty would influence the load prediction accuracy. Major conclusions are twofold: first, LSTM performs well in short-term prediction (1 h ahead) but not in long term prediction (24 h ahead), because the sequential information becomes less relevant and accordingly not so useful when the prediction horizon is long. Second, the presence of weather forecast uncertainty deteriorates XGBoost’s accuracy and favors LSTM, because the sequential information makes the model more robust to input uncertainty. Training the model with the uncertain rather than accurate weather data could enhance the model’s robustness. Our findings have two implications for practice. First, LSTM is recommended for short-term load prediction given that weather forecast uncertainty is unavoidable. Second, XGBoost is recommended for long term prediction, and the model should be trained with the presence of input uncertainty.
Understanding the relationship between occupant behaviors and building energy consumption is one of the most effective ways to bridge the gap between predicted and actual energy consumption in ...buildings. However effective methodologies to remove the impact of other variables on building energy consumption and isolate the leverage of the human factor precisely are still poorly investigated. Moreover, the effectiveness of statistical and data mining approaches in finding meaningful correlations in data is largely undiscussed in literature. This study develops a framework combining statistical analysis with two data-mining techniques, cluster analysis and association rules mining, to identify valid window operational patterns in measured data. Analyses are performed on a data set with measured indoor and outdoor physical parameters and human interaction with operable windows in 16 offices. Logistic regression was first used to identify factors influencing window opening and closing behavior. Clustering procedures were employed to obtain distinct behavioral patterns, including motivational, opening duration, interactivity and window position patterns. Finally the clustered patterns constituted a base for association rules segmenting the window opening behaviors into two archetypal office user profiles for which different natural ventilation strategies as well as robust building design recommendations that may be appropriate. Moreover, discerned working user profiles represent more accurate input to building energy modeling programs, to investigate the impacts of typical window opening behavior scenarios on energy use, thermal comfort and productivity in office buildings.
•Data mining is a powerful tool to identify behavioral patterns in building data sets.•Clustering is employed to obtain distinct window operation behavioral patterns.•Motivational, opening duration, interactivity and window position patterns are highlighted.•Association rules identify archetypal user profiles in offices of the same building.•Discerned user profiles represent robust inputs to building energy modeling programs.
•Developed methods and used data models to integrate city’s public building records.•Shading from neighborhood buildings strongly influences urban building performance.•A case study demonstrated the ...workflow, simulation and analysis of building retrofits.•CityBES retrofit analysis feature provides actionable information for decision making.•Discussed significance and challenges of urban building energy modeling.
Buildings in cities consume 30–70% of total primary energy, and improving building energy efficiency is one of the key strategies towards sustainable urbanization. Urban building energy models (UBEM) can support city managers to evaluate and prioritize energy conservation measures (ECMs) for investment and the design of incentive and rebate programs. This paper presents the retrofit analysis feature of City Building Energy Saver (CityBES) to automatically generate and simulate UBEM using EnergyPlus based on cities’ building datasets and user-selected ECMs. CityBES is a new open web-based tool to support city-scale building energy efficiency strategic plans and programs. The technical details of using CityBES for UBEM generation and simulation are introduced, including the workflow, key assumptions, and major databases. Also presented is a case study that analyzes the potential retrofit energy use and energy cost savings of five individual ECMs and two measure packages for 940 office and retail buildings in six city districts in northeast San Francisco, United States. The results show that: (1) all five measures together can save 23–38% of site energy per building; (2) replacing lighting with light-emitting diode lamps and adding air economizers to existing heating, ventilation and air-conditioning (HVAC) systems are most cost-effective with an average payback of 2.0 and 4.3years, respectively; and (3) it is not economical to upgrade HVAC systems or replace windows in San Francisco due to the city’s mild climate and minimal cooling and heating loads. The CityBES retrofit analysis feature does not require users to have deep knowledge of building systems or technologies for the generation and simulation of building energy models, which helps overcome major technical barriers for city managers and their consultants to adopt UBEM.
•Three zoning methods were presented, evaluated and compared in the UBEM context.•Influences of using a floor multiplier in urban building energy modeling are studied.•940 office and retail buildings ...in San Francisco were simulated with 3 zoning methods.•Modeling each floor as one zone underestimates thermal loads and equipment capacity.•Zoning methods have a significant impact on the simulated energy use of UBEM.
Urban-scale building energy modeling (UBEM)—using building modeling to understand how a group of buildings will perform together—is attracting increasing attention in the energy modeling field. Unlike modeling a single building, which will use detailed information, UBEM generally uses existing building stock data consisting of high-level building information. This study evaluated the impacts of three zoning methods and the use of floor multipliers on the simulated energy use of 940 office and retail buildings in three climate zones using City Building Energy Saver. The first zoning method, OneZone, creates one thermal zone per floor using the target building’s footprint. The second zoning method, AutoZone, splits the building’s footprint into perimeter and core zones. A novel, pixel-based automatic zoning algorithm is developed for the AutoZone method. The third zoning method, Prototype, uses the U.S. Department of Energy’s reference building prototype shapes. Results show that simulated source energy use of buildings with the floor multiplier are marginally higher by up to 2.6% than those modeling each floor explicitly, which take two to three times longer to run. Compared with the AutoZone method, the OneZone method results in decreased thermal loads and less equipment capacities: 15.2% smaller fan capacity, 11.1% smaller cooling capacity, 11.0% smaller heating capacity, 16.9% less heating loads, and 7.5% less cooling loads. Source energy use differences range from -7.6% to 5.1%. When comparing the Prototype method with the AutoZone method, source energy use differences range from -12.1% to 19.0%, and larger ranges of differences are found for the thermal loads and equipment capacities. This study demonstrated that zoning methods have a significant impact on the simulated energy use of UBEM. One recommendation resulting from this study is to use the AutoZone method with floor multiplier to obtain accurate results while balancing the simulation run time for UBEM.
This paper takes a close look at the China national standard GB50189-2005, Design Standard for Energy Efficiency of Public Buildings, which was enforced on July 1, 2005. The paper first reviews the ...standard, then compares the standard with ASHRAE Standard 90.1-2004 to identify discrepancies in code coverage and stringency, and recommends some energy conservation measures that can be evaluated in the design of public buildings to achieve energy savings beyond the standard. The paper also highlights several important features of 90.1-2004 that may be considered as additions to the GB50189-2005 standard during the next revision. At the end the paper summarizes the latest developments in building energy standards and rating systems in China and the US.
•Occupant behavior has significant impacts on energy use in buildings.•A simulation approach is proposed to estimate energy savings of behavior measures.•Five measures on lighting, plug-loads, HVAC, ...windows, and thermostat are studied.•Behavior measures can achieve up to 41.0% savings based on the simulation results.•Occupancy schedule significantly affects the energy savings of behavior measures.
Occupant behavior in buildings is a leading factor influencing energy use in buildings. Low-cost behavioral solutions have demonstrated significant potential energy savings. Estimating the behavioral savings potential is important for a more effective design of behavior change interventions, which in turn will support more effective energy-efficiency policies. This study introduces a simulation approach to estimate the energy savings potential of occupant behavior measures. First it defines five typical occupant behavior measures in office buildings, then simulates and analyzes their individual and integrated impact on energy use in buildings. The energy performance of the five behavior measures was evaluated using EnergyPlus simulation for a real office building across four typical U.S. climates and two vintages. The Occupancy Simulator was used to simulate the occupant movement in each zone with inputs from the site survey of the case building. Based on the simulation results, the occupant behavior measures can achieve overall site energy savings as high as 22.9% for individual measures and up to 41.0% for integrated measures. Although energy savings of behavior measures would vary depending upon many factors, the presented simulation approach is robust and can be adopted for other studies aiming to quantify occupant behavior impact on building performance.
•Overview of city's public buildings datasets for urban building energy modeling.•Developed methods to integrate city building datasets using common standards.•Simplified, standardized and produced ...the building dataset in CityGML, GeoJSON, and FileGDB formats.•A case study demonstrated the integration of diverse datasets from multiple city departments.•Discussed key challenges of data integration for city buildings.
Urban building energy modeling (UBEM) is becoming a proven tool to support energy efficiency programs for buildings in cities. Development of a city-scale dataset of the existing building stock is a critical step of UBEM to automatically generate energy models of urban buildings and simulate their performance. This study introduces data needs, data standards, and data sources to develop city building datasets for UBEM. First, a literature review of data needs for UBEM was conducted. Then, the capabilities of the current data standards for city building datasets were reviewed. Moreover, the existing public data sources from several pioneer cites were studied to evaluate whether they are adequate to support UBEM. The results show that most cities have adequate public data to support UBEM; however, the data are represented in different formats without standardization, and there is a lack of common keys to make the data mapping easier. Finally, a case study is presented to integrate the diverse data sources from multiple city departments of San Francisco. The data mapping process is introduced and discussed. It is recommended to use the unique building identifiers as the common keys in the data sources to simplify the data mapping process. The integration methods and workflow are applied to other U.S. cities for developing the city-scale datasets of their existing building stock, including San Jose, Los Angeles, and Boston.
Building simulation: Ten challenges Hong, Tianzhen; Langevin, Jared; Sun, Kaiyu
Building simulation,
10/2018, Letnik:
11, Številka:
5
Journal Article
Recenzirano
Odprti dostop
Buildings consume more than one-third of the world’s primary energy. Reducing energy use and greenhouse-gas emissions in the buildings sector through energy conservation and efficiency improvements ...constitutes a key strategy for achieving global energy and environmental goals. Building performance simulation has been increasingly used as a tool for designing, operating and retrofitting buildings to save energy and utility costs. However, opportunities remain for researchers, software developers, practitioners and policymakers to maximize the value of building performance simulation in the design and operation of low energy buildings and communities that leverage interdisciplinary approaches to integrate humans, buildings, and the power grid at a large scale. This paper presents ten challenges that highlight some of the most important issues in building performance simulation, covering the full building life cycle and a wide range of modeling scales. The formulation and discussion of each challenge aims to provide insights into the state-of-the-art and future research opportunities for each topic, and to inspire new questions from young researchers in this field.
•A stochastic modeling method with OB models to predict district loads is proposed.•This method was applied to a case study in China using the DeST simulation engine.•This study summarizes the ...typical occupancy schedules and OB modes in HSCW of China.•The proposed method performs better than two conventional simulation methods.•Applications and limitations of the proposed method are also discussed.
District cooling systems are widely used in urban residential communities in China. Most of such systems are oversized, which leads to wasted investment, low operational efficiency and, thus, waste of energy. The accurate prediction of district cooling loads that can support the rightsizing of cooling plant equipment remains a challenge. This study develops a novel stochastic modeling method that consists of (1) six prototype house models representing most apartments in a district, (2) occupant behavior models of residential buildings reflecting their spatial and temporal diversity as well as their complexity based on a large-scale residential survey in China, and (3) a stochastic sampling process to represent all apartments and occupants in the district. The stochastic method was applied to a case study using the Designer’s Simulation Toolkit (DeST) to simulate the cooling loads of a residential district in Wuhan, China. The simulation results agreed well with the measured data based on five performance metrics representing the aggregated cooling consumption, the peak cooling loads, the spatial load distribution, the temporal load distribution and the load profiles. Two prevalent simulation methods were also employed to simulate the district cooling loads. The results showed that oversimplified assumptions about occupant behavior could lead to significant overestimation of the peak cooling load and the total cooling loads in the district. Future work will aim to simplify the workflow and data requirements of the stochastic method for its application, and to explore its use in predicting district heating loads and in commercial or mixed-use districts.