Search

In an era of increasing energy demands and environmental concerns, optimizing energy consumption within buildings is crucial. Despite the vast improvements in HVAC and lighting systems, plug loads remain an under-studied area for enhancing building energy efficiency. This paper studies smart plug active operating mode detection, plug-level load forecasting, and plug scheduling methodologies. This research leverages a unique dataset from the University of California, San Diego, consisting of readings from over 150 smart plugs in several office buildings for more than a year, notably during the post-Covid era. This dataset is made publicly available. A comprehensive literature review on plug, i.e. , appliances operating mode detection is presented. Novel unsupervised learning approaches are applied to identify plug operating modes. A pipeline integrating the detected modes with forecasting and scheduling is developed, aiming at building energy consumption reduction. Our findings offer valuable insights and promising results into smart plug management for energy-efficient buildings. • A novel year-long dataset of power readings from 169 office smart plugs is introduced. • The first review of plug load mode detection methods and terminology is presented. • The paper proposes an unsupervised method to identify plugs load operating modes. • The proposed smart plug scheduling pipeline reduces energy consumption by 20%. [ABSTRACT FROM AUTHOR]

With a rising number of buildings being equipped with private distributed energy resources (DERs) such as rooftop PV panels and energy storage devices, an effective energy management method for a building cluster becomes increasingly imperative. This paper proposes a novel decentralized transactive energy management (TEM) method for a hybrid cluster of residential and commercial buildings, which enables peer-to-peer (P2P) energy trading among the DER owners and consumers. The strategic interactions among the DER owners and consumers are modeled as a multi‑leader-multi-follower (MLMF) Stackelberg game and formulated as a bi-level model. The DER owners, the commercial and residential consumers are all autonomous entities, optimizing their individual welfare functions and sharing necessary trading-related information, which are expressed as the upper-level leaders' models and the lower-level followers' models, respectively. To preserve the privacy and autonomy of each entity within the building cluster, a distributed algorithm incorporated with an efficient P2P pricing mechanism is designed for the formulated MLMF Stackelberg game model. Simulation results demonstrate the effectiveness of the proposed method on mitigating the reliance of the building cluster on the power grid, motivating the DERs to actively participate in P2P trading, and reducing the consumers' energy consumption costs • A decentralized transactive energy management framework is designed for a real-world hybrid building cluster. • The strategic interactions among the owners of distributed energy resources (DERs) and the consumers within the building cluster are captured by a multi‑leader-multi-follower Stackelberg game. • A novel distributed algorithm is developed to protect the privacy and autonomy of participants in a large-scale building cluster. [ABSTRACT FROM AUTHOR]

In the transition from fossil fuel to electrified heating, a concerning trend is emerging in certain regions of the US. Owners of buildings with gas-based systems leave them in place after adding heat pumps (HPs). Existing control solutions for these hybrid (dual fuel) systems are rudimentary and fall short of realizing the full carbon reduction potential of these systems. Model predictive control (MPC) is often regarded as the benchmark for achieving optimal control in integrated systems. However, in the case of small-medium commercial buildings (SMCBs), the control and communication infrastructure required to facilitate the implementation of such advanced controls is often lacking. This paper presents a field implementation of easy-to-deploy MPC for a dual fuel heating system consisting of HPs and a gas-fired furnace (GF) for SMCBs. The control system is deployed on an open-source middleware platform and utilizes low-cost sensor devices to be used for real SMCBs without major retrofits. We demonstrated this MPC in a real office building with 5 HPs and 1 GF for 2 months. The test results showed that MPC reduced 27% of cost while completely eliminating GF usage by shifting 23% of the thermal load from occupied-peak time to non-occupied-non-peak times. • Implementation of a practical MPC for a heat pump-based dual fuel system in a small commercial building. • Experimental demonstration of a practical MPC with low-cost sensor retrofits for 2 winter months. • Achievement of 27% energy cost saving and 23% load shifting from occupied-peak time to non-occupied-non-peak time. • Elimination of natural gas usage with GHG emissions reduction of ∼ 52 kgCO2/month. [ABSTRACT FROM AUTHOR]

• An online platform for dynamic lifecycle analysis is presented. • An energy-retrofitting use case in a multi-family residential building is examined. • The primary energy needs, carbon emissions, and lifecycle costs are analyzed. • The retrofitting use case can lead to a 95 % reduction of CO2 eq emissions. • For 25 years of operation, a total of 515 k€ can be saved. Building stock retrofitting is essential to achieve the ambitious sustainability goals of the building sector due to its high energy consumption rates. The evaluation of the various building interventions shall be holistically assessed in terms of environmental and costing impact. The aim of this paper is twofold: First, it presents the innovative characteristics of a developed online tool (Virtual intEgrated platfoRm on LIfe cycle AnalYsis - VERIFY) able to perform dynamic life cycle analysis and global warming impact assessments by capitalizing on the well-known LCA and LCC methodologies, applicable in the case of building renovation. VERIFY is able to analyse dynamic life cycle inventories that consider the temporal profiles of energy consumption, and the time-dependent temperature changes, while being also interoperable in terms of exchanging data with other available energy simulation engines, or even using real-time monitoring data from sensors, processing any data time granulation. Second, the paper evaluates, from a life cycle perspective, the impact of specific energy retrofitting measures, meeting the Passive House Standard, for the case of a multi-family residential building in Athens, Greece. The proposed energy-retrofitting scenario examines actions related to the deep retrofitting of the building envelope and the upgrade of the thermal components as well as to the incorporation of clean electricity generation based on renewable energy systems; all aiming to drastically reduce the environmental impact of the building, rendering it almost near zero energy. Through the planned infrastructure installations, the primary energy needs and CO 2eq emissions were reduced by 91 % and by 95 % respectively, while for a building operational lifespan of 25 years, savings up to 515 k€ compared to the baseline scenario, can be achieved. [ABSTRACT FROM AUTHOR]

With global warming intensifying and resource conflicts escalating, the world is undergoing a transformative shift toward sustainable practices and energy-efficient solutions. With more than 32% of the global energy used by commercial and residential buildings, there is an urgent need to revisit traditional approaches to Building Energy Management (BEM). Within a BEMSplatform, regulating the operation of Heating, Ventilation, and Air Conditioning (HVAC) systems is more important, noting that HVAC systems account for about 40% of the total energy cost in the commercial sector. This paper offers a Deep Reinforcement Learning (DRL) algorithm as a data-driven approach to controlling HVAC operation to enhance the energy efficiency of commercial buildings with open offices while ensuring thermal comfort for occupants in different zones. Compared to alternative methods such as rule-based models and model-predictive control, data-driven models have shown promising results in optimizing building energy consumption without the need for building-specific thresholds, prior knowledge about the underlying physics of heat distribution, and digital mapping of the airflow. Despite the astonishing performance of modern DRL methods in controlling energy management, a particular energy-saving solution for open-plan offices with multiple Variable Air Volume (VAV) systems, where different zones cannot be treated independently, is still missing. Also, some of the existing methods suffer from key issues such as long training time and lack of generalizability for using over-complicated models, incorporating external factors that are hard to model and characterize, and including factors that are not typically accessible. To solve these issues, we propose a low-complexity DRL-based model with multi-input multi-output architecture for the HVAC energy optimization of open-plan offices, which uses only a handful of controllable and accessible factors. The efficacy of our solution is evaluated through extensive analysis of the overall energy consumption and thermal comfort levels compared to a baseline system based on the existing HVAC schedule from a real case. This comparison shows that our method achieves 37% savings in energy consumption with minimum temperature violation (< 1%) of the desired temperature range during work hours. It takes only a total of 40 min for 5 epochs (about 7.75 min per epoch) to train a network with superior performance and covering diverse conditions for its low-complexity architecture; therefore, it easily adapts to changes in the building setups, weather conditions, occupancy rate, etc. Moreover, by enforcing smoothness on the control strategy, we suppress the frequent and unpleasant on/off transitions on HVAC units to avoid occupant discomfort and potential damage to the system. The generalizability of our model is verified by applying it to different building models and under various weather conditions. [Display omitted] • Model and analyze the thermodynamics of multi-zone open-plan offices • DRL-based control algorithm to optimize thermal comfort and energy efficiency • 37% reduction in HVAC energy with less than 1% temperature violation • Smoothness-imposing reward function to minimize on-off transitions • Applicable to different floor plans under various weather conditions [ABSTRACT FROM AUTHOR]

• A data-driven model was developed to rapidly quantify the demand response potential of building HVAC systems. • Multiple factors affecting demand response potential were considered in the data-driven model. • A co-simulation platform based on EnergyPlus and Modelica was built for fully getting the demand response potential data. • Two commonly used DR strategies were evaluated using a series of indices. Buildings and heating, ventilation and air-conditioning systems are appropriate resources for demand response, but there is still a lack of rapid quantification methods to reveal buildings' dynamic demand response potential under different meteorological conditions and control strategies. In order to achieve real-time power grid dispatch and to facilitate grid-building interaction, this paper proposes a framework for establishing models to rapidly quantify the demand response potential of resources. First, a detailed co-simulation model of the building and its energy system is constructed, followed by extensive stochastic simulations of two commonly used demand response strategies, after which the demand response potential is quantified using a set of indices. Then, feature selection is performed on factors that may affect demand response potential, and a data-driven model for quantifying demand response potential is proposed, whose results are compared with those from the simulation model. Also, the demand response effects of the two control strategies are compared in this paper. A case study was carried out in an industrial building in Shenzhen, China. The results showed that under different boundaries, the coefficient of determination (R 2) of the quantification indices are all above 0.9, and that the data-driven model rapidly and accurately captured the demand response potential of the building and its heating, ventilation and air-conditioning system, and was easy to deploy in practice. [ABSTRACT FROM AUTHOR]

Hemp-based building envelopes have gained significant popularity in developed countries, and now the trend of constructing houses with hemp-clay blocks is spreading to developing countries like Morocco. Investigating the hygrothermal behavior of such structures under actual climate conditions is essential for advancing and promoting this sustainable practice. This paper presents an in-depth experimental characterization of a commercial hemp-clay brick that has been exposed to the outdoor environment for four years, in addition to field measurements on a building scale demonstration prototype. Additionally, the study simulates 17 representative cities to assess the hygrothermal performance and energy-saving potential in each of Morocco's six existing climate zones, using the EnergyPlus engine. The experimental campaign's findings demonstrate excellent indoor air temperature and relative humidity regulation within the hemp-clay wall building, leading to satisfactory levels of thermal comfort within hemp-clay wall buildings. This is attributed to the material's good thermal conductivity and excellent moisture buffering capacity (found to be 0.31 W/mK and 2.25 g/m2%RH), respectively). The energy simulation findings also point to significant energy savings, with cooling and heating energy reductions ranging from 27.7% to 47.5% and 33.7% to 79.8%, respectively, as compared to traditional Moroccan buildings. [Display omitted] • Experimental characterization of thermal conductivity, sorption isotherm, and moisture buffering capacity of hemp bricks. • Assessment of indoor thermal comfort and energy demand in a hemp-based building across 17 cities in Morocco. • Hemp-clay bricks demonstrated excellent humidity regulation with a moisture buffer capacity of 2.25 g/m2%RH. • Hemp-clay walls can achieve energy savings ranging from 27.7%–47.5% for cooling and 33.7%–79.8% for heating in Morocco. [ABSTRACT FROM AUTHOR]

The optimization of the operation of heating, ventilation, and air conditioning (HVAC) systems has gained considerable attention due to its energy conservation potential. Collecting various building operation data presents an opportunity to improve the efficiency of energy management and reduce consumption. In this paper, a data-driven online energy management framework is proposed for enhancing the effectiveness of HVAC systems and minimizing energy consumption. To streamline the modeling process and improve the accuracy of temperature estimation, our proposed framework adopts a symbolic regression approach to estimate building thermodynamics based on the collected data. It also formulates a model predictive control (MPC)-based HVAC scheduling strategy to optimize HVAC operations while maximizing thermal comfort and reducing energy consumption and peak power demand. We conducted on-site experiments to evaluate the proposed framework and demonstrated a notable reduction in total energy consumption in both HVAC cooling and heating operations: average reductions of 4.9% and 30.2%, and 49.3% and 73.9% for best cases in the on-site experiment under certain conditions. In the heating operation, the proposed data-driven HVAC management framework efficiently utilized indoor/outdoor heat generation to maintain comfortable temperature levels without activating the HVAC. Such effective utilization of heat sources shortened the HVAC runtimes and further reduced the total energy consumption. With the proposed framework, the average HVAC peak power demand fell by 25.8% and 35.1% during cooling and heating operations in the on-site experiment with certain scenarios. On-site experiment results demonstrated the proposed framework can efficiently suppress peak power demand, reduce energy consumption in both cooling and heating operation under certain conditions. The framework's data-driven approach provides an effective means of optimizing HVAC operations, reducing energy consumption, and improving the thermal comfort of the building's occupants. • Propose a data-driven online energy management framework for HVAC systems. • Employ a symbolic regression approach for estimating building thermodynamics. • Model predictive control approach is employed to achieve real-time energy management. • First on-site expeiment for data-driven HVAC management in office environment. • The proposed framework reduce the energy consumption up to 73.9%. [ABSTRACT FROM AUTHOR]

Integrated energy system (IES) is characterized by high self-consumption ratio of on-site generated renewable energy, high efficiency of conventional energy utilization and possesses a significant flexibility in its operation. This overall, constitute to the foundation of low-carbon communities. Considering economic and environmental benefits, this paper proposes a two-layer co-optimization model with the upper layer optimizing the IES configuration and the lower layer optimizing IES operation. A community in Beijing is introduced as a case study to analyze the benefits of IES and compared with the conventional energy system. Multiple scenarios are researched, including (a) IES for low-carbon communities with PV roof area limitation, (b) IES for each of the four building types with PV roof area limitation, (c) IES for low-carbon communities without PV roof area limitation. The results indicate that the optimized IES for community can reduce the cost by 25% and CO 2 emissions by 32% per year by considering the limitation of roof area for the PV system. Without this constraint, the costs and emissions could be reduced by 20% and 62%, respectively. In addition, IES is more appropriate for office and commercial buildings since their load characteristics allow for load shifting and community-level IES costs and emissions are 8% and 10% lower than building-level due to the complementary of community load. This study provides suggestions for the IES planning and application in low-carbon community. [Display omitted] • An IES with multiple renewable energy and energy storage subsystems is proposed. • A two-layer co-optimization model of IES design and operation is developed. • IES optimization results for building-level and community-level are compared. • Impact of roof area limitation for PV on IES optimization results is investigated. [ABSTRACT FROM AUTHOR]

• A novel system is presented that predicts day-ahead demand profile and optimizes EV charging/discharging to minimize peak demand. • The system is designed to comply with preplanned EV trip schedules and minimum state of charge (SOC) requirements. • Four machine learning methods are tested to predict day ahead demand profile of a case study building at 15-minute intervals. • Five peak demand reduction scenarios are analyzed using combination of EVs, a stationary battery, and a PV system. • The results show up to 36 % reduction in peak demand using two EVs, one stationary battery, and PV system of 40 kW capacity. Reducing electricity peak demand is essential to maintain the balance between supply and demand side in electricity power markets, as well as reduce utility costs and environmental impacts. With growth in adoption of Electric Vehicles (EVs), there is an emerging opportunity to balance electrical power demand of buildings by storing electricity in EVs during low demand periods and discharging electricity into buildings during peak demand periods. Due to uncertainty in time and magnitude of peak demand, decision makers are always faced with a challenging task to identify optimal schedules for charging and discharging EVs to minimize peak electricity demand. This paper presents the development of a novel system that is capable of predicting day-ahead building electricity demand profile and identifying optimum schedule of charging and discharging EVs to minimize electricity peak demand. The system is designed to comply with planned EV trip schedules and minimum state of charge (SOC). The system consists of (1) machine Learning (ML) model to predict electrical power demand, and (2) demand management optimization model to identify optimal schedule for charging and discharging EVs. Four methods are explored to develop the ML model, including histogram-based gradient boosting, random forest, deep artificial neural network (DNN), and long short-term memory (LSTM). A case study of multi-tenant commercial building is analyzed to evaluate the performance of the system and demonstrate its new capabilities. The results of the case study shows that LSTM has the best performance in terms of mean absolute error, root mean square error, and mean absolute percentage error with average values of 7.44, 17.78, and 20.08 %, respectively. Five scenarios for shaving peak electricity demand, including combinations of two electric vehicles, a stationary battery, and a PV system are investigated. Scenarios including the stationary battery and the PV system are considered to evaluate the full potential of peak demand reduction in the case study building. The results of the demand management optimization model show up to 36 % reduction in peak demand using two EVs, one stationary battery, and PV system of 40 kW capacity. The key contributions that this study adds to existing knowledge are: (1) developing machine learning models to predict day-ahead electricity demand in 15-minute intervals, (2) integrating machine learning and optimization algorithms in identifying EV charging and discharging schedules to minimize utility cost by shaving peak energy demand, and (3) considering planned EV trips and minimum SOC requirements in identifying optimal charging and discharging of EVS to shave peak energy demand in buildings. The implementation of this system provides practical solutions for managing electricity demand in commercial buildings using EVs. By reducing energy consumption and promoting the innovative use of EVs, this system offers a sustainable approach to managing electricity demand in commercial buildings. [ABSTRACT FROM AUTHOR]

• Machine Learning is known as a reliable technique to predict the energy performance of buildings. • The accuracy, training, and prediction speed of 24 ML models are evaluated and compared in this study. • Top-ten most accurate ML models in predicting PV self-consumption are introduced by the obtained results. • The TES system is an effective solution for improving PV self-consumption in buildings. • The LCC-based optimum size of TES leads to 7.1% saving in the total life cycle cost of the building case study. The application of Photovoltaic (PV) system in buildings is growing rapidly in response to the need for clean energy sources and building decarbonization targets. Nonetheless, enhancing PV self-consumption through technical solutions such as Energy Storage Systems (ESS) is getting higher importance to increase the profitability of PV plants, by minimizing the building-grid interaction. In this context, analyzing PV self-consumption of different energy storage configurations becomes more relevant and crucial in building energy modeling although it is heavily time-consuming and complicated, particularly within a multi-objective optimization related to the ESS design. As a solution to resolve this issue, this paper evaluates the accuracy, training, and prediction speed of 24 Machine Learning (ML) models to be used as surrogate models for analyzing PV self-consumption in smart buildings. Furthermore, the performance of short-term Thermal Energy Storage (TES) to increase PV self-consumption is assessed and presented using ML models. The results showed the Gaussian Process Regression (GPR), Neural Networks (NN) including bilayered and trilayered NN models, Support Vector Machines (SVM) including the fine gaussian and cubic SVM models, and Ensembles of Trees (EoT) as superior ML models. The results also revealed that TES systems can efficiently increase PV self-consumption in the building equipped with electric heat pumps to provide heating, cooling, and domestic hot water. Moreover, the TES size optimization regarding the Life Cycle Cost (LCC) showed that the LCC-based optimum TES size can yield 7.1% savings within 30 years of the building service life. The novelties of this research are first to provide a reference to select the most suitable ML models in predicting PV self-consumption, second to implement machine learning for analyzing the performance of short-term thermal energy storage to enhance PV self-consumption in buildings, and third to carry out an LCC-based optimization on TES size using ML-based prediction models. [ABSTRACT FROM AUTHOR]

• We used quantum computing to optimize model predictive control of building HVAC systems. • We formulated mixed-integer non-linear programming as quadratic unconstrained binary optimization for quantum computer. • Solution of quantum computing was almost the same as traditional optimization method, but computing time was greatly reduced. • Quantum computing had the potential to solve large-scale non-linear optimization problems for building energy systems. Buildings contribute to more than 70% of overall U.S. electricity usage and greenhouse gas (GHG) emissions. HVAC systems in buildings often consume more than 40% of the total building energy usage. To reduce such high energy use, numerous control strategies including optimal and predictive controls have been developed and demonstrated. To achieve a near real-time solution, most previous research has simplified the non-linearity of building thermodynamics and provided an approximate optimal solution. The future HVAC control optimizes more connected devices in buildings, which requires a rapid and accurate response, not only to the building itself but also to the grid signals. It also poses the challenge of solving non-linear problems with discrete variables. With the recent development of quantum computers, this has become feasible. In this paper, we developed a new optimization solution based on quantum annealing for model predictive control (MPC) of a rooftop unit (RTU). Compared to traditional optimization methods, we obtained similar solutions with less than 2% differences and improved computational speed from hours to seconds. We also demonstrated an 80% reduction in total electricity consumption and a 21% reduction in electricity bills by considering day-ahead price time-of-use demand response signals. Quantum computing has proven capable of solving large-scale non-linear discrete optimization problems for building energy systems. [ABSTRACT FROM AUTHOR]

• An energy system optimization framework considering ZEB conditions and EV charging demand is proposed. • Machine learning models and Energyplus simulation are incorporated to predict building energy demand. • SGA2 and PROBID are applied to solve the multi-objective optimization problem. • EV charging demand can impact energy system design. • ZEB criteria can be the economic buffer to achieve zero carbon emissions. With increasing concerns over carbon dioxide emissions, the concept of Zero Energy Building (ZEB) has emerged. Electric Vehicles (EVs) are also considered environmentally friendly since they reduce greenhouse gas emissions, with a rapidly growing market. With these global trends of increasing EV amount and infrastructure, building energy systems should incorporate the ZEB concept and the increasing electricity requirements for EV charging. However, it is unclear how EV charging demand can affect building energy system design while aligning with ZEB requirements. Therefore, this paper develops a new framework to find the optimal energy system design that meets EV charging demand and ZEB requirements. The charging demand for EVs is predicted by the machine learning model, which combines the building energy demand from EnergyPlus. Ultimately, the Genetic Algorithm and PROBID method are applied to optimize the Total Annual Cost (TAC) and Self-Energy Sufficiency Ratio. EV charging demand has been found to affect energy system design, especially in small-size buildings. Using the proposed method, the building owner can determine the optimal capacity of an energy system based on economic and ZEB conditions, contributing to the future net ZEB and transportation systems. [ABSTRACT FROM AUTHOR]

• Analyzing various architectural function ratio combinations of urban complex. • Evaluating energy performance of urban complex with various function combinations. • Providing function scheme data set of urban complex in terms of energy performance. • Improving pre-programming process of urban complex by considering energy factor. In the context of the global energy shortage, urban complexes have attracted more and more attention from energy-saving research due to their large number and great energy-saving potential. While, the energy performance factors are rarely considered in the function scheme decision of urban complex in the recent pre-programming process. Optimizing the building area ratio scheme of urban complex is the key link to improve the comprehensive energy utilization efficiency. This paper proposes an approach to optimize building area ratio scheme of urban complex from energy performance perspective by energy simulation, scenario analysis and demand side management methods, and provides a data set of rankings of urban complex function combination schemes by evaluating the performance of energy load leveling and the degree of load stability of each building area ratio scheme in different climatic conditions, respectively. The results indicate that: 1) the different building area ratios notably affect the integrated operating energy performance of urban complexes with a maximum evaluation indexes difference of more than 85%; 2) the evaluating and optimizing results of the building area ratio schemes show broadly similarity in different climate conditions; 3) the optimal building area ratio scheme of urban complex is usually with less "shopping mall" (0 ∼ 10%), more "residence" (more than 35%), and floating ratio of "office" and "hotel" (0 ∼ 35%). The main findings of this study provide a quantitative basis for the architectural pre-programming and function scheme decision-making of urban complex by demand-side energy management to improve the energy efficiency, so as to guarantee the energy supply security. [ABSTRACT FROM AUTHOR]

This paper introduces a simple strategy for modular building energy management with explicit demand response based on a three-level hierarchical model predictive control that engages the entire building in optimal operation and financially viable flexibility provision. Financial viability of the assessed flexibility capacity is guaranteed regardless of the flexibility activation scenario, under the assumption that the entire building flexibility capacity is accepted. The strategy is verified on three pilot buildings. The selected pilot buildings have diverse configurations and are tested under different conditions to show the broad applicability of the developed approach. The analysis of the building flexibility is focused on particular characteristic days in the heating and cooling season and provides a comparison of the overall building operation costs for the following three control options applied to the building: hierarchical control with and without flexibility provision as well as conventional control. In this way, it is possible to quantify the benefits achievable exactly due to the advanced energy management system operation on the site. Flexibility sources and amounts within the building are analyzed in detail for each specific case. • Hierarchical MPC-based BEMS for operation scheduling and flexibility provision. • Assessment of the entire building flexibility capacity. • Guaranteed financial viability of flexibility provision. • Verification with a detailed analysis on 3 diverse case-study buildings. [ABSTRACT FROM AUTHOR]

[Display omitted] • A physical-model-free coordination control of multi-buildings is proposed. • Novel multi-agent deep reinforcement learning framework for continuous control. • The generalization of developed method is verified on consecutive datasets. • Developed algorithm informs effective and fast control against uncertainties. • Developed method is superior to 4 advanced benchmark algorithms. In commercial buildings, Heat, Ventilation, and Air Conditioning (HVAC) systems account for about 40–50 % of total electricity usage, contributing to an economic burden on building operators. Furthermore, increasing amounts of distributed generation can bring challenges and opportunities for voltage regulation in low voltage distribution networks. In this paper, we tend to develop intelligent management to save the electricity bills of HVAC systems in multiple multi-zone buildings while relieving the stress of voltage regulation across the network. However, it is challenging to achieve the above aims due to the existence of parameter uncertainties (e.g., electricity load, outdoor temperature, photovoltaic generation, etc.), a sizeable continuous decision space, unknown thermal dynamics model, and distribution network topology, and a non-convex multi-objective function. In this context, a novel model-free multi-agent deep reinforcement learning (MADRL)-based multi-building control algorithm is proposed to achieve building-side and grid-level objectives. The proposed method adopts a centralized training and decentralized execution framework while integrating an attention mechanism to ease training and preserve privacy. This also enables the agent to achieve control purposes based only on local measurements, reducing communication cost. Simulation results based on real-world data verify that the proposed method can achieve real-time physical-model-free control of multi-buildings to tackle fast fluctuations of voltage and temperature caused by the uncertain external factors while being advantageous over other two MADRL methods. Additionally, comparison analysis on untouched datasets illustrates that the proposed method achieves similar results and better computation performance with the perfect physical-model-based approach. [ABSTRACT FROM AUTHOR]

• Design a techno-economic EMS that optimizes the operating costs. • Ensure stability and power quality using a flat controller. • Provide an uninterruptible power supply (UPS) to the load. • Improve system efficiency with the higher possible battery SoC. An optimal techno-economic multi-level energy management strategy (EMS) is designed for a DC microgrid applied to a commercial building. The proposed power system is a grid-connected microgrid composed of a fuel cell (FC) system, a photovoltaic (PV) array, and a battery storage unit. The interaction between various types of power sources may lead to stability problems. Moreover, providing an uninterruptible and high-quality power supply, reducing operating costs, and maximizing system efficiency are essential factors. This paper presents a techno-economic EMS, one of the most significant objectives for the following years. The proposed EMS comprises two levels for minimizing the operating cost and supplying the load and another control level based on flatness control theory to improve the power quality. The first level is based on one of the following EMSs: State Machine Control (SMC), Equivalent Consumption Minimization Strategy (ECMS), and External Energy Maximization Strategy (EEMS) to improve efficiency and reduce the required power. The second level is based on the economic dispatching algorithm (EDA) to reduce the operating cost and ensure a UPS. Finally, the system performance is elaborated; the simulation results prove the superiority of the EEMS-EDA combination, where the cost-saving can be enhanced by 3.47%, and the efficiency is about 80.44%. [ABSTRACT FROM AUTHOR]

• A discrete Bayesian Network (DisBN) method is developed to diagnose cross-level faults in HVAC systems. • Weather and schedule-based pattern matching (WPM) method is used to generate the baseline and part of DisBN parameters. • System operation interval data is collected to demonstrate the effectiveness of the developed WPM-DisBN method. • The root causes of cross-level faults can be diagnosed and isolated to address the shortcomings of component-level fault diagnostic solutions. Fault detection and diagnosis (FDD) technologies are critical to ensure satisfactory building performance, such as reducing energy wastes and negative impacts on occupant comfort and productivity. Existing FDD technologies mainly focus on component-level FDD solutions, which could lead to mis-diagnosis of cross-level faults in heating, ventilating, and air-conditioning (HVAC) systems. Cross-level faults are those faults that occur in one component or subsystem, but cause operational abnormalities in other components or subsystems, and result in a building level performance degradation. How to effectively diagnose the root cause of a cross-level fault is the focus of this study. This paper presents a novel discrete Bayesian Network (DisBN)-based method for diagnosing cross-level faults in an HVAC system commonly used in commercial buildings. A two-level DisBN structure model is developed in this study. The parameters used in the DisBN model are obtained either from expert knowledge or through machine-learning strategies from normal system operation data. Meanwhile, the probability parameters are discretized to incorporate the uncertainties associated with typical expert knowledge. Thus, the developed DisBN method addresses the challenges many other BN based FDD methods face, i.e., the lack of fault data for BN parameter training. The developed DisBN represents causal relationships between a fault and its cross-level system impacts (i.e., fault symptoms or fault indicators) by considering how fault impacts propagate across different levels in an HVAC system. A weather and schedule information-based Pattern Matching (WPM) method is employed to automatically create WPM baseline data sets for each incoming real time snapshot data from the building systems. Consequently, BN inference and real-time diagnostics are achieved by comparing incoming snapshot data and the WPM baseline data set. The proposed method is evaluated using experimental fault data collected in a campus building. Fault diagnosis results demonstrate that the WPM-DisBN method is effective at locating the root causes of cross-level faults in an HVAC system. [ABSTRACT FROM AUTHOR]

Home energy management systems (HEMS) have been shown to reduce energy bills and to provide grid services including peak demand reduction and demand flexibility. However, uncertainty in residential energy systems is a significant issue and can reduce the benefits of a HEMS to the homeowner or grid operator. Sources of uncertainty include weather forecasts, predictions of energy-related occupant activities (e.g., hot water draws), and parameter estimation for the building envelope and energy-consuming equipment. This paper tackles the problem of uncertainty by developing a framework that simulates HEMS in uncertain conditions and evaluates the performance of multiple control strategies. A linear, reduced-order residential building model for model predictive control applications is derived and compared to a full-order model. Stochastic model predictive control is shown to perform better than deterministic and heuristic methods when considering realistic forecasts with uncertainty. The framework can evaluate the performance of HEMS in real-world applications, which can help de-risk HEMS deployment. • Uncertainty from weather and occupant forecasts impact home energy management. • A stochastic control framework is developed for a home energy management system. • A reduced-order linear house model is developed for model predictive control. • Incorporating uncertain forecasts reduces control performance in HEMS. • Stochastic MPC outperforms common deterministic and heuristic control methods. [ABSTRACT FROM AUTHOR]

Model predictive control (MPC) has shown potential in improving building performance but is bottlenecked by the difficulty in constructing control-oriented models. The challenge lies in evaluating the sufficiency of the model and the data usage beforehand. This paper bridges the knowledge gaps in the interactions between data requirements, model quality, and control performance by integrating real-world measurements and simulation-based experiments. The data usage related to occupancy and Internal heat gain (IHG) was studied considering its importance and the absence of consensus in the literature. Varying occupant-related data sources were tested as RC model inputs, including none, schedule, electricity consumption, CO 2 ppm, and ideal measurement. Combinations of model inputs and complexities were examined for prediction and control in an office, a classroom, and multi-zone offices on one floor. The results indicated that the usefulness of data is jointly affected by three factors: measurement suitability, model complexity, and modeling purpose. Given the adequate model structure, satisfying prediction and control performance was achieved in offices with no detailed measurement. Meanwhile, electricity and CO 2 were needed together to capture the IHG influence and realize the good performance for classrooms. The experiments also uncovered the heterogeneous requirements on models from traditional prediction tests and the control tasks. Lower prediction error did not always mean better control. More importantly, we provided the first quantitative demonstration of the complementary relationship between model adequacy and data informativeness with respect to different purposes. This study advocates the pioneering idea of sparse data usage and parsimonious modeling, which promotes the actual application of MPC in buildings by guiding control-oriented model development. [Display omitted] • Over 10% energy saving and acceptable PMV achieved without occupant-related data. • Occupant-related data is needed only if the internal heat gain has high variability. • Adequate model with sufficient data yields physically representative identification. • Better physical representation could lead to larger RMSE but more precise control. • Sparse data and parsimonious modeling should be pursued in real-world implementation. [ABSTRACT FROM AUTHOR]

• Hybrid bottom-up model to integrate geo-referenced and synthetic stock models. • Demonstrate multi-scale and long-term transitional workflow for commercial building stock. • Identify drivers and determinants of energy usage in the commercial building stock. • 32% cumulative performance gap if concurrent consideration of physical and technical factors is disregarded. • Introducing the unit block concept to determine minimum viable requirements. For energy modelling of commercial building stock (CBS), several methodologies were developed focusing on technology alternatives and retrofits for climate change mitigation, but they often have limitations to involve a differentiated description of practices and policies across a range of scales and sectors. This paper presents a novel hybrid model by integrating spatial and synthetic modelling approaches to facilitate the concurrent consideration of multiple building-oriented elements. The model is further implemented on Japanese building stock to demonstrate multi-scale and long-term framework for energy modelling of CBS. The results indicate the following findings: the scale choice introduces significant variability in energy consumption estimation with a relative error of 10–17% mainly depending on building geometry and plug loads; the long-term model exhibits a higher scale-bounded uncertainty in terms of energy saving potential that varies by 2–3 times when a non-representative scale is applied; disregarding physical and technical factors results in a cumulative performance gap of up to 32%; and the accuracy and extent of the model across the stock are more sensitive to technical factors, such as building system stock dynamics and the heterogeneity of HVAC systems, than physical factors. The model provides a multi-tier framework using spatial intelligence building stock approach to develop long-term energy efficiency monitoring strategies for CBS. [ABSTRACT FROM AUTHOR]

• The optimal energy benefits of switchable insulation systems (SIS) are evaluated for office buildings. • When combined with precooling, optimally operated SISs can provide significant energy and cost reductions. • Optimal and hybrid controls are developed to operate SIS-integrated roofs and precooling strategies. • Optimized controls for SIS and precooling can reduce by over 20% energy use and 30% peak demand. In this paper, optimized controls to enhance the energy efficiency and demand response benefits of roof-integrated switchable insulation system combined with precooling strategies are evaluated for office buildings. A wide range of optimal cost functions are investigated to evaluate the total building energy saving as well as electrical peak demand shedding and shifting potentials. The results indicate that roof-integrated switchable insulation systems provide modest reductions even when optimally controlled of up to 5% and 4% savings in total building's energy use and electrical peak demand during a peak day. However, when precooling strategies are integrated with roof switchable insulation systems, optimal controls can enhance the performance of office buildings by reducing total energy use by up to 19% and electrical peak demand by up to 34%. Moreover, the analysis presented in this study demonstrated that roof-integrated SIS combined with precooling strategies offer significant potential to shape the energy demand profiles for grid-interactive efficient building (GEB) applications when optimized controls are deployed. [ABSTRACT FROM AUTHOR]

• A new design approach is presented for selecting smart glazing for office buildings. • Optimal optical properties for smart glazing are determined for office spaces. • Impacts of daylighting are considered to identify optimal optical properties. • Optimal smart glazing can achieve over 40% annual energy savings for office spaces. In this paper, a systematic analysis is presented to determine optimal optical properties of smart glazing to minimize energy demand for perimeter office spaces. The optimization analysis considers the impacts of design and operation conditions of commercial buildings including size and orientation of windows as well as the climate characteristics of the buildings. In addition, daylighting benefits are included in the analysis of smart glazing that can transition between clear and dark states as well as intermediate tint levels. The optimization analysis results confirm that smart glazing can provide substantial energy efficiency for office buildings when coupled with daylighting controls especially for cold climates and large windows. However, the analysis indicates that specific set of optical properties for both clear and dark states are required to achieve optimal energy efficiency benefits for smart glazing when applied to office spaces. Indeed, the presented analysis allows the identification of the optimal optical properties for smart glazing depending on several influencing factors including primarily the climate as well as the window size and orientation. Moreover, the magnitude of the annual energy savings achieved by the smart glazing depends largely on the optical properties of the static glazing type used as a reference. Using code-compliant static glazing as a reference, the energy savings incurred by smart glazing range from 1.2% to 13.3% depending on daylighting controls and climatic conditions. [ABSTRACT FROM AUTHOR]

Despite extensive empirical evidence of the environmental benefits of green buildings and the increasing urgency to reduce carbon emissions in cities, there has been limited widespread adoption of energy retrofit investments in existing buildings. In this paper, we empirically model financial returns to energy retrofit investments for more than 3600 multifamily and commercial buildings in New York City, using a comprehensive database of energy audits and renovation work extracted from city records using a natural language processing algorithm. Based on auditor cost and savings estimates, the median internal rate of return for adopted energy conservation measures is 21% for multifamily buildings and 25% for office properties. Logistic regression modeling demonstrates adoption rates are higher for office buildings than multifamily, and in both cases adopter buildings tend to be larger, higher value, and less energy efficient prior to retrofit implementation. The economically significant magnitudes of returns to adopted energy conservation measures raise important questions about why many property owners choose not to adopt. As such, we discuss incentive and regulatory mechanisms that can overcome financial and informational barriers to the adoption of energy efficiency measures. • A large-scale empirical study of financial returns to building energy retrofits. • A novel dataset of audit recommendations and permitted work developed using NLP. • Analysis of IRR and NPV curves for both adopter and non-adopter buildings. • Median IRRs are found to be 21% for multifamily and 25% for office buildings. • Retrofit decisions not fully explained by potential investment returns. [ABSTRACT FROM AUTHOR]

• Increasingly complex machine learning models lead to lower interpretability. • An attention-based neural network architecture for load prediction is proposed. • The architecture is tested using real operational data from a commercial building. • Proposed architecture shows enhanced prediction accuracy and interpretability. Machine learning has gained increasing popularity in building energy management due to its powerful capability and flexibility in model development as well as the rich data available in modern buildings. While machine learning is becoming more powerful, the models developed, especially artificial neural networks like Recurrent Neural Networks (RNN), are becoming more complex, resulting in "darker models" with lower model interpretability. The sophisticated inference mechanism behind machine learning prevents ordinary building professionals from understanding the models, thereby lowering trust in the predictions made. To address this, attention mechanisms have been widely implemented to improve the interpretability of deep learning; these mechanisms enable a deep learning-based model to track how different inputs influence outputs at each step of inference. This paper proposes a novel neural network architecture with an attention mechanism for developing RNN-based building energy prediction, and investigates the effectiveness of this attention mechanism in improving the interpretability of RNN models developed for 24-hour ahead building cooling load prediction. To better understand, explain and evaluate these neural network-based building energy prediction models, the obtained attention vectors (or metric) are used to visualize the influence of different parts of model inputs on the prediction result. This helps the users to understand why predictions are made by the model, as well as how input sequences proportionally influence the output sequences. Further analysis of attention vectors can provide interesting temporal information for understanding building thermal dynamics, like the thermal inertia of the building. The proposed attention-based architecture can be implemented in developing optimal operation control strategies and improving demand and supply management. The model developed based on this architecture is assessed using real building operational data, and shows improved accuracy and interpretability over baseline models (without adopting attention mechanisms). The research results help to bridge the gap between building professionals and advanced machine learning techniques. The insights obtained can be used as guidance for the development, fine-tuning, explanation and debugging of data-driven building energy prediction models. [ABSTRACT FROM AUTHOR]