An agent-based stochastic Occupancy Simulator

Occupancy has significant impacts on building performance. However, in current building performance simulation programs, occupancy inputs are static and lack diversity, contributing to discrepancies between the simulated and actual building performance. This paper presents an Occupancy Simulator that simulates the stochastic behavior of occupant presence and movement in buildings, capturing the spatial and temporal occupancy diversity. Each occupant and each space in the building are explicitly simulated as an agent with their profiles of stochastic behaviors. The occupancy behaviors are represented with three types of models: (1) the status transition events (e.g., first arrival in office) simulated with probability distribution model, (2) the random moving events (e.g., from one office to another) simulated with a homogeneous Markov chain model, and (3) the meeting events simulated with a new stochastic model. A hierarchical data model was developed for the Occupancy Simulator, which reduces the amount of data input by using the concepts of occupant types and space types. Finally, a case study of a small office building is presented to demonstrate the use of the Simulator to generate detailed annual sub-hourly occupant schedules for individual spaces and the whole building. The Simulator is a web application freely available to the public and capable of performing a detailed stochastic simulation of occupant presence and movement in buildings. Future work includes enhancements in the meeting event model, consideration of personal absent days, verification and validation of the simulated occupancy results, and expansion for use with residential buildings.     

Simulation and visualization of energy-related occupant behavior in office buildings

In current building performance simulation programs, occupant presence and interactions with building systems are over-simplified and less indicative of real world scenarios, contributing to the discrepancies between simulated and actual energy use in buildings. Simulation results are normally presented using various types of charts. However, using those charts, it is difficult to visualize and communicate the importance of occupants’ behavior to building energy performance. This study introduced a new approach to simulating and visualizing energy-related occupant behavior in office buildings. First, the Occupancy Simulator was used to simulate the occupant presence and movement and generate occupant schedules for each space as well as for each occupant. Then an occupant behavior functional mockup unit (obFMU) was used to model occupant behavior and analyze their impact on building energy use through co-simulation with EnergyPlus. Finally, an agent-based model built upon AnyLogic was applied to visualize the simulation results of the occupant movement and interactions with building systems, as well as the related energy performance. A case study using a small office building in Miami, FL was presented to demonstrate the process and application of the Occupancy Simulator, the obFMU and EnergyPlus, and the AnyLogic module in simulation and visualization of energy-related occupant behaviors in office buildings. The presented approach provides a new detailed and visual way for policy makers, architects, engineers and building operators to better understand occupant energy behavior and their impact on energy use in buildings, which can improve the design and operation of low energy buildings.     

A generalised stochastic model for the simulation of occupant presence

This paper describes an algorithm for the simulation of occupant presence, to be later used as an input for future occupant behaviour models within building simulation tools. By considering occupant presence as an inhomogeneous Markov chain interrupted by occasional periods of long absence, the model generates a time series of the state of presence (absent or present) of each occupant of a zone, for each zone of any number of buildings. Tested on occupancy data from private offices, the model has proven its capacity to realistically reproduce key properties of occupant presence such as times of arrival and departure, periods of intermediate absence and presence as well as periods of long absence from the zone. This model (due to related metabolic heat gains), and associated behavioural models which use occupants’ presence as an input, have direct consequences for building energy consumption.     

Evaluating diverse patterns of occupant behavior regarding control-based activities in energy performance simulation

Simulation is recognized as an effective tool for building energy performance assessment during design orretrofit processes. Nevertheless, simulation models yield deviating outcomes from the actual building performance and a significant portion of this deviation originates from the dynamic nature of occupant behavior. Literature on occupant behavior indicates that occupant behavior is not integrated into building energy performance assessment procedures with appropriate resolution, instead they are acceptedas as sumedand fixed data sets that usually represent the presence of occupants. This study attempts to evaluate the effect of diverse patterns of occupant behavior on energy performance simulation for office buildings. Diverse levels of sensitivity of occupant behavior on control-based activities such as using lighting apparatus, adjusting thermostat settings, and presence in space are employed through three diverse occupant behavior patterns. These occupancy patterns are correlated with three identical office spaces simulated within a conceptual office building. EDSL Tas is used to run building energy performance simulations. Effects of occupant behavior patterns on simulation outcomes are compared for five sample winter and summer workdays, with respect to heating and cooling loads. Results present findings on how diversity of occupancy profiles influences the consumption outcomes.     

Buildings.Occupants: a Modelica package for modelling occupant behaviour in buildings

Energy-related occupant behaviour is crucial to design and operation of energy and control systems in buildings. Occupant behaviours are often oversimplified as static schedules or settings in building performance simulation ignoring their stochastic nature. The continuous and dynamic interaction between occupants and building systems motivates their simultaneous simulation in an efficient manner. In the past, simultaneous simulation has relied on co-simulation approaches or customized source code changes to building simulation programmes. This paper presents Buildings. Occupants, an open-source package implemented in Modelica, for the simulation of occupant behaviours of lighting, windows, blinds, heating and air conditioning systems in office and residential buildings. Examples were presented to illustrate how the models in the Occupants package are capable to simulate stochastic occupant behaviours. The major contribution of this work is to introduce the equation-based modelling approach to simulate occupant behaviours in buildings and to develop an open-source Occupants package in the Modelica language.     

Modeling occupancy and behavior for better building design and operation—A critical review

People spend more than 90% of their life time in buildings, which makes occupant behavior one of the leading influences of energy consumption in buildings. Occupancy and occupant behavior, which refer to human presence inside buildings and their active interactions with various building system such as lighting, heating, cooling, ventilation, window blinds, and plugs, attract great attention of research with regard to better building design and operation. Due to the stochastic nature of occupant behavior, prior occupancy models vary dramatically in terms of data sampling, spatial and temporal resolution. This paper provides a comprehensive review of the current modeling efforts of occupant behavior, summarizes occupancy models for various applications including building energy performance analysis, building architectural and engineering design, intelligent building operations and building safety design, and presents challenges and areas where future research could be undertaken. In addition, modeling requirement for different applications is analyzed. Furthermore, a few commonly used statistical and data mining models are presented. The purpose of this paper is to provide a modeling reference for future researchers so that a proper method or model can be selected for a specific research purpose.     

Lighting energy consumption in ultra-low energy buildings: Using a simulation and measurement methodology to model occupant behavior and lighting controls

As building owners, designers, and operators aim to achieve significant reductions in overall energy consumption, understanding and evaluating the probable impacts of occupant behavior becomes a critical component of a holistic energy conservation strategy. This becomes significantly more pronounced in ultra-efficient buildings, where system loads such as heating, cooling, lighting, and ventilation are reduced or eliminated through high-performance building design and where occupant behavior-driven impacts reflect a large portion of end-use energy. Further, variation in behavior patterns can substantially impact the persistence of any performance gains. This paper describes a methodology of building occupant behavior modeling using simulation methods developed by the Building Energy Research Center (BERC) at Tsinghua University using measured energy consumption data collected by the University of Washington Integrated Design Lab (UW IDL). The Bullitt Center, a six-story 4831 m² (52,000 ft²) net-positive-energy urban office building in Seattle, WA, USA, is one of the most energy-efficient buildings in the world (2013 WAN Sustainable Building of the Year Winner). Its measured energy consumption in 2015 was approximately 34.8 kWh/(m²?yr) (11 kBtu/(ft²?yr)). Occupant behavior exerts an out-sized influence on the energy performance of the building. Nearly 33% of the end-use energy consumption at the Bullitt Center consists of unregulated miscellaneous electrical loads (plug-loads), which are directly attributable to occupant behavior and equipment procurement choices. Approximately 16% of end-use energy is attributable to electric lighting which is also largely determined by occupant behavior. Key to the building’s energy efficiency is employment of lighting controls and daylighting strategies to minimize the lighting load. This paper uses measured energy use in a 330 m² (3550 ft²) open office space in this building to inform occupant profiles that are then modified to create four scenarios to model the impact of behavior on lighting use. By using measured energy consumption and an energy model to simulate the energy performance of this space, this paper evaluates the potential energy savings based on different occupant behavior. This paper describes occupant behavior simulation methods and evaluates them using a robust dataset of 15 minute interval sub-metered energy consumption data. Lighting control strategies are compared via simulation results, in order to achieve the best match between occupant schedules, controls, and energy savings. Using these findings, we propose a simulation methodology that incorporates measured energy use data to generate occupant schedules and control schemes with the ultimate aim of using simulation results to evaluate energy saving measures that target occupant behavior.     

Energy and comfort performance of thermally activated building systems including occupant behavior

In building simulations it is common practice to use standardized occupant behavior and internal gains. Although this is a valid approach for designing systems, the probabilistic nature of these boundary conditions influences the energy demand and achieved thermal comfort of real systems. This paper analyzes the influence of occupant behavior on the energy performance and thermal comfort of a typical office floor equipped with a thermally activated building system (TABS). A multi-zone TRNSYS model with 10 adjacent zones per orientation for a typical moderate Belgian climate is set up. First, the energy performance and thermal comfort of thermally activated building systems (TABS) are compared with the performance of idealized cooling with standardized user behavior. TABS are able to deliver good thermal comfort but show to have a higher energy demand. Secondly, probabilistic occupant behavior was implemented in the TABS simulations. The influence of the occupancy rate, the shading device use and switching of the lights are analyzed by defining user profiles. It is shown that occupant behavior may have an important influence on the cooling demand and thermal comfort. However, as long as good solar protection is foreseen and operated in a correct way, TABS are able to cope with different user behavior modeled in this paper. In this case, normal daily stochastic processes do not considerably affect the cooling demand and thermal comfort during summer.     

An information technology enabled sustainability test-bed (ITEST) for occupancy detection through an environmental sensing network

This paper describes a large-scale wireless and wired environmental sensor network test-bed and its application to occupancy detection in an open-plan office building. Detection of occupant presence has been used extensively in built environments for applications such as demand-controlled ventilation and security; however, the ability to discern the actual number of people in a room is beyond the scope of current sensing techniques. To address this problem, a complex sensor network is deployed in the Robert L. Preger Intelligent Workplace comprising a wireless ambient-sensing system, a wired carbon dioxide sensing system, and a wired indoor air quality sensing system. A wired camera network is implemented as well for establishing true occupancy levels to be used as ground truth information for deriving algorithmic relationships with the environment conditions. To our knowledge, this extensive and diverse ambient-sensing infrastructure of the ITEST setup as well as the continuous data-collection capability is unprecedented. Final results indicate that there are significant correlations between measured environmental conditions and occupancy status. An average of 73% accuracy on the occupancy number detection was achieved by Hidden Markov Models during testing periods. This paper serves as an exploration to the research of ITEST for occupancy detection in offices. In addition, its utility extends to a wide variety of other building technology research areas such as human-centered environmental control, security, energy efficient and sustainable green buildings.     

Bayesian estimation of occupancy distribution in a multi-room office building based on CO<Subscript>2</Subscript> concentrations

Occupancy information in an office building is an important asset for determining energy-efficient operations and emergency evacuation of a building. In this study, we developed a method to estimate the occupancy distribution in a multi-room office building using Bayesian inference. The Markov chain Monte Carlo algorithm was used to estimate the real-time occupancy in individual rooms based on indoor carbon dioxide concentrations. The office building was made-up of five rooms with different physical configurations and dimensions, and the rooms were air-conditioned and ventilated by a central air handling unit. The carbon dioxide concentration data were generated by the simulation software CONTAMW according to a given schedule of occupancy and the supply airflow rates in each room. The objective of the present paper is to investigate the effects of various parameters of Bayesian inference on the accuracy of estimation results. The parameters include the probability of prior information, the uncertainty level of CO₂ data, and the time interval for monitoring CO₂.     

Statistical analysis and modeling of occupancy patterns in open-plan offices using measured lighting-switch data

Occupancy profile is one of the driving factors behind discrepancies between the measured and simulated energy consumption of buildings. The frequencies of occupants leaving their offices and the corresponding durations of absences have significant impact on energy use and the operational controls of buildings. This study used statistical methods to analyze the occupancy status, based on measured lighting-switch data in five-minute intervals, for a total of 200 open-plan (cubicle) offices. Five typical occupancy patterns were identified based on the average daily 24-hour profiles of the presence of occupants in their cubicles. These statistical patterns were represented by a one-square curve, a one-valley curve, a two-valley curve, a variable curve, and a flat curve. The key parameters that define the occupancy model are the average occupancy profile together with probability distributions of absence duration, and the number of times an occupant is absent from the cubicle. The statistical results also reveal that the number of absence occurrences decreases as total daily presence hours decrease, and the duration of absence from the cubicle decreases as the frequency of absence increases. The developed occupancy model captures the stochastic nature of occupants moving in and out of cubicles, and can be used to generate a more realistic occupancy schedule. This is crucial for improving the evaluation of the energy saving potential of occupancy based technologies and controls using building simulations. Finally, to demonstrate the use of the occupancy model, weekday occupant schedules were generated and discussed.     

Occupant behaviour simulation for cellular offices in early design stages—Architectural and modelling considerations

Building simulation is most useful and most difficult in early design stages. Most useful since the optimisation potential is large and most difficult because input data are often not available at the level of resolution required for simulation software. The aim of this paper is to addresses this difficulty, by analysing the predominantly qualitative information in early stages of an architectural design process in search for indicators towards quantitative simulation input. The discussion in this paper is focused on cellular offices. Parameters related to occupancy, the use of office equipment, night ventilation, the use of lights and blinds are reviewed based on simulation input requirements, architectural considerations in early design stages and occupant behaviour considerations in operational stages. A worst and ideal case scenario is suggested as a generic approach to model occupant behaviour in early design stages when more detailed information is not available. Without actually predicting specific occupant behaviour, this approach highlights the magnitude of impact that occupants can have on comfort and building energy performance and it matches the level of resolution of available architectural information in early design stages. This can be sufficient for building designers to compare the magnitude of impact of occupants with other parameters in order to inform design decisions. Potential indicators in early design stages towards the ideal or worst case scenario are discussed.     

Evaluation of distributed environmental control systems for improving IAQ and reducing energy consumption in office buildings

Conventional heating, ventilation, and air conditioning (HVAC) systems are incapable of providing control over individual environments or adjusting fresh air supply based on the dynamic occupancy of individual rooms in an office building. This paper introduces the concept of distributed environmental control systems (DECS) and shows that improvement in indoor air quality (IAQ) and energy efficiency can be achieved by providing required amounts of fresh air directly to the individual office spaces through distributed demand controlled ventilation (DDCV). In DDCV, fresh air is provided to each micro-environment (room or cubicle) based on input from distributed sensors (CO₂, VOC, occupancy, etc.) or intelligent scheduling techniques to provide acceptable IAQ for each occupant, rather than for groups or populations of occupants. In order to study DECS, a numerical model was developed that incorporates some of the best available models for studying building energy consumption, indoor air flow, contaminant transport and HVAC system performance. The developed model was applied to a DECS in a model office building equipped with a DDCV system. By implementing DECS/DDCV and intelligent scheduling techniques it is possible to achieve an improvement in IAQ along with a reduction in annual energy consumption compared to conventional ventilation systems.     

Effect of thermostat and window opening occupant behavior models on energy use in homes

Existing dynamic energy simulation tools exceed the static dimension of the simplified methods through a better and more accurate prediction of energy use; however, their ability to predict real energy consumption is undermined by a weak representation of human interactions with the control of the indoor environment. The traditional approach to building dynamic simulation considers energy consumption as fully deterministic, taking into account standardized input parameters and using fixed and unrealistic schedules (lighting level, occupancy, ventilation rate, thermostat set-point). In contrast, in everyday practice occupants interact with the building plant system and building envelope in order to achieve desired indoor environmental conditions. In this study, occupant behavior in residential building was modelled accordingly to a probabilistic approach. A new methodology was developed to combine probabilistic user profiles for both window opening and thermostat set-point adjustments into one building energy model implemented in the dynamic simulation tool IDA Ice. The aim of the study was to compare mean values of the probabilistic distribution of the obtained results with a singular heating energy consumption value obtained by means of standard deterministic simulations. Major findings of this research demonstrated the weakness of standardized occupant behavior profile in energy simulation tools and the strengths of energy models based on measurements in fields and probabilistic modelling providing scenarios of occupant behavior in buildings.     

A model for assessing occupant flow in building spaces

As buildings are becoming larger in size, the need to efficiently plan and predict the occupant movement in building spaces is gaining more attention and importance. Occupant movement in large public buildings such as train stations, airports, universities, hospitals, and shopping centers needs to be carefully analyzed and predicted for safety and also for other issues such as the level of service (LOS), comfort, and short-term planning. Currently there are many detailed occupant/pedestrian simulation models that can predict the level of service in corridors and other dynamic spaces in the building in terms of occupant flow. However, these simulation models require a well-formulated design with detailed design features. In addition, significant investment in time and effort is required in order to build the models for simulation analysis. Therefore, there is a need for a simple and quick analysis method to aid in the sizing and design of building spaces during the early design stages so that these spaces can accommodate occupant flow efficiently and safely. This paper presents a method to evaluate the LOS of occupants in dynamic buildings spaces without the need for building and running detailed simulations, so that designers can understand how well a particular space accommodates occupants' movements and activities early on in the design phase. The proposed method can be used to determine the occupant flow density in a wide array of building layouts and designs as it correlates to the level of service. A mathematical model that offers a closed-form formula for sizing space for occupant flow is developed. The model is presented in this paper and validated using real-life data. Results should be of interest to practicing architects as well as researchers.     

Smart building management vs. intuitive human control—Lessons learnt from an office building in Hungary

Smart building management and control are adopted nowadays to achieve zero-net energy use in buildings. However, without considering the human dimension, technologies alone do not necessarily guarantee high performance in buildings. An office building was designed and built according to state-of-the-art design and energy management principles in 2008. Despite the expectations of high performance, the owner was facing high utility bills and low user comfort in the building located in Budapest, Hungary. The objective of the project was to evaluate the energy performance and comfort indices of the building, to identify the causes of malfunction and to elaborate a comprehensive energy concept. Firstly, current building conditions and operation parameters were evaluated. Our investigation found that the state-of-the-art building management system was in good conditions but it was operated by building operators and occupants who are not aware of the building management practice. The energy consumption patterns of the building were simulated with energy modelling software. The baseline model was calibrated to annual measured energy consumption, using actual occupant behaviour and presence, based on results of self-reported surveys, occupancy sensors and fan-coil usage data. Realistic occupant behaviour models can capture diversity of occupant behaviour and better represent the real energy use of the building. This way our findings and the effect of our proposed improvements could be more reliable. As part of our final comprehensive energy concept, we proposed intervention measures that would increase indoor thermal comfort and decrease energy consumption of the building. A parametric study was carried out to evaluate and quantify energy, comfort and return on investment of each measure. It was found that in the best case the building could save 23% of annual energy use. Future work includes the follow-up of: occupant reactions to intervention measures, the realized energy savings, the measurement of occupant satisfaction and behavioural changes.     

Identification of key factors for uncertainty in the prediction of the thermal performance of an office building under climate change

There is growing concern about the potential impact of climate change on the thermal performance of buildings. Building simulation is well-suited to predict the behaviour of buildings in the future, and to quantify the risks for prime building functions like occupant productivity, occupant health, or energy use. However, on the time scales that are involved with climate change, different factors introduce uncertainties into the predictions: apart from uncertainties in the climate conditions forecast, factors like change of use, trends in electronic equipment and lighting, as well as building refurbishment / renovation and HVAC (heating, ventilation, and air conditioning) system upgrades need to be taken into account. This article presents the application of two-dimensional Monte Carlo analysis to an EnergyPlus model of an office building to identify the key factors for uncertainty in the prediction of overheating and energy use for the time horizons of 2020, 2050 and 2080. The office has mixed-mode ventilation and indirect evaporative cooling, and is studied using the UKCIP02 climate change scenarios. The results show that regarding the uncertainty in predicted heating energy, the dominant input factors are infiltration, lighting gain and equipment gain. For cooling energy and overheating the dominant factors for 2020 and 2050 are lighting gain and equipment gain, but with climate prediction becoming the one dominant factor for 2080. These factors will be the subject of further research by means of expert panel sessions, which will be used to gain a higher resolution of critical building simulation input.     

Assessing the total energy impact of manual and optimized blind control in combination with different lighting schedules in a building simulation environment

In this work, we investigated the interplay and the influence of lighting and blind control models on the heating, cooling, and lighting energy loads of an office room. By including different stochastic models for occupancy and appliances, we built a complete simulation environment based on the building simulation program IDA ICE. For control models, we implemented different strategies, including a simple on/off scheme, a realistic model of occupants, and an optimized control. In literature, the results are often compared with simple on/off schemes, which are not hard to beat in terms of performance. With an optimal control, the real saving potential is assessed, which can be used as benchmark case for comparison with other control models. Results based on annual simulations show that active occupants can reduce energy consumption by up to 50% from a worst-case scenario, whereas advanced controllers can further reduce the consumption by another 60%.     

EnergyPlus model-based predictive control within design–build–operate energy information modelling infrastructure

This study proposes a design–build–operate energy information modelling (DBO-EIM) infrastructure to allow users to deploy the design-stage building energy model for model predictive control (MPC) system in the building operation. A newly constructed office building is studied as a test bed. An EnergyPlus model-based predictive control (EPMPC) system is designed and simulated in the Matlab/Simulink environment within the DBO-EIM infrastructure. EPMPC aims at minimizing heating, ventilation, and air conditioning energy consumption while maintaining occupant thermal comfort. Compared to the baseline rule-based control system, EPMPC demonstrates a 28.9% energy reduction in one-week simulation in the heating season and 2.7% energy reduction in one-week simulation in the cooling season. The comfort constraint is met during more than 90% of the simulated hours. The study demonstrates one significant contribution of the DBO-EIM infrastructure that a design-stage EnergyPlus model can be integrated in an MPC system and the preliminary simulation results are satisfactory.