Energy consumption in office buildings highly depends on occupant energy-use behaviors and intervening these behaviors could function as a cost-effective approach to enhance energy savings. Current behavior-intervention techniques extensively rely on occupant-specific energy-use information at the workstation level and often ignore shared appliances. It is because an occupant typically has full responsibility for her workstation appliances energy consumption and shares the responsibility of the shared appliances energy consumption. However, understanding energy-use behavior of both workstation and shared appliances is necessary for applying appropriate behavior-intervention techniques. Despite this importance, there is still no practical and scalable method to capture personalized energy-use information of workstation and shared appliances since the conventional methods use plug-in power meters that are extremely expensive and difficult to maintain over long period of time. To address this gap, we propose a comprehensive occupant-level energy-usage approach which utilizes the data from the internet of things devices in office buildings to provide information related to energy-use behavior of workstation and shared appliances of each occupant in an economical and feasible manner. In particular, we introduce an energy behavior index which quantitatively compares individual occupants’ energy-consuming data to identify high energy consumers and inefficient behaviors. Results from an experiment conducted in an office building equipped with internet of things devices demonstrate the feasibility of the proposed approach to classify occupants to different energy-usage categories. Our proposed approach along with appropriate behavior-intervention techniques could be used to impact occupant energy-use behaviors.

Wi-Fi networks are currently considered as an efficient and economical tool for occupancy sensing in office buildings. Studies particularly indicated that these networks could be utilized to understand/predict occupants’ energy-use patterns. Despite the value that investigating this possibility could provide for the current research, it has not been well explored how energy-use pattern information could be extracted from Wi-Fi system information. In response, this study utilizes statistical analyses to investigate the correlation of Wi-Fi flows with miscellaneous electric loads (MELs) in office buildings. MELs account for more than one-third of office-building energy consumption and are the best representative of occupants’ energy-use patterns. In the pursuit of the objective, data from two offices were collected over a 3-month period of time. Results from the analyses show that an average 92 percent of MELs energy consumption could be predicted through the Wi-Fi flows in a building. This finding thereby demonstrates that occupants’ energy-use patterns are highly positively correlated to Wi-Fi flows in a building and accordingly, the information of Wi-Fi networks could be utilized to understand/interpret these patterns. This significantly contributes to the current body of research and can be used to support efforts into understanding/enhancing occupants’ energy-use behaviors. In addition, since Wi-Fi networks are a major subset of internet of things (IoT) hardware systems and IoT implementation for intelligent energy management in buildings significantly depends on occupant energy-use patterns, this research helps IoT-based efforts by displaying how these patterns could be extracted from IoT infrastructure.

We introduce a systematic approach for analyzing the energy consumption of four control policies (i.e., zone level daily optimal control, zone level annual optimal control, building level daily optimal control, building level annual optimal control), which differed based on their temporal and spatial control scales. In order to integrate occupant thermal comfort requirements, we defined uniformly distributed random constraint functions on the setpoints. We used the DOE reference small office building in three U.S. climate zones for simulating the performances of control policies, using EnergyPlus. Among the four control policies, the building level annual control policy showed close to the highest energy efficiency (27.76% to 50.91% (average of 39.81%) savings depending on the climate) with comparatively small training data requirements. In addition, the building level daily optimal setpoint selection, subject to thermal comfort constraints, leads to 17.64 – 38.37% (average of 26.61%) energy savings depending on the climate. We also demonstrate that temporal scale of the policies have a statistically significant impact on the small office building’s energy consumption while spatial scale’s impact is insignificant.

HVAC systems are the major energy consumers in commercial buildings in the United States. These systems are operated to provide comfortable thermal conditions for building occupants. The common practice of defining operational settings for HVAC systems is to use fixed set points, which assume occupants have static comfort requirements. However, thermal comfort has been shown to vary from person to person and also change over time due to climatic variations or acclimation. In this paper, we introduce an online learning approach for modeling and quantifying personalized thermal comfort. In this approach, we fit a probability distribution to each comfort condition (i.e., uncomfortably warm, comfortable, and uncomfortably cool) data set and define the overall comfort of an individual through combing these distributions in a Bayesian network. In order to identify comfort variations over time, Kolmogorov–Smirnov test is used on the joint probability distributions. In order to identify comfortable environmental conditions, a Bayesian optimal classifier is trained using online learning. In order to validate the approach, we collected data from 33 subjects, and an average accuracy of 70.14% and specificity of 76.74% were achieved. In practice, this approach could transform the comfort objectives to constrain functions and prevents pareto optimality problems.

The common practice of defining operational settings for Heating, Ventilation and Air Conditioning (HVAC) systems in buildings is to use fixed set points, which assume occupants have same and static comfort requirements. However, thermal comfort varies from person to person and also changes due to climatic variations or acclimation, making it dynamic. In addition, thermal comfort in transient conditions are different from the steady state conditions, which makes the prediction of thermal comfort more difficult. Thus, thermal comfort has to be monitored over time. In this paper, we present a novel infrared thermography based technique to monitor an individual’s thermoregulation performance and thermal comfort levels by measuring the skin temperature on several points on human face, which has a high density of blood vessels and is not usually covered by clothing. Unlike other methods, our method requires no continuous user input or interaction. Our results demonstrate that the monitored facial points behave differently under the heat and cold stresses and it can be explained based on the underlying vascular territories. We define two heuristics to describe the thermoneutral zone based on the observed behaviors and estimate thermal comfort for individuals with 95% confidence level. Considerable variations are observed in the thermoregulation performance and uncomfortably cool conditions metrics between the males and females. Females’ thermoregulation system responses are less sensitive to the perception of warm conditions. However, similar behaviors are observed for uncomfortably cool conditions across genders.

Maintaining thermal comfort in built environments is important for occupant health, well-being, and productivity, and also for efficient HVAC system operations. Most of the existing personal thermal comfort learning methods require occupants to provide feedback via a survey to label the monitored environmental or physiological conditions in order to train the prediction models. However, the accuracy of these models usually drops after the training process as personal thermal comfort is dynamic and changes over time due to climatic variations and/or acclimation. In this paper, we present a hidden Markov model (HMM) based learning method to capture personal thermal comfort using infrared thermography of human face. We chose human face since its blood vessels has a higher density and it is not covered while performing regular activities in built environments. Learning algorithm has 3 hidden states (i.e., uncomfortably warm, comfortable, uncomfortably cool) and uses discretization for forming the observed states from the continuous infrared measurements. The approach can potentially be used for continuous monitoring of thermal comfort to capture the variations over time. We tested and validated the method in a four-day long experiment with 10 subjects and demonstrated an accuracy of 82.8% for predicting uncomfortable conditions.

Previous research efforts for optimizing energy usage of HVAC systems require either mathematical models of HVAC systems to be built or they require substantial historical operational data for learning the optimal operational settings. We introduce a model-free control policy that begins learning optimal settings with no prior historical data and optimizes HVAC operations. The control policy is an adaptive hybrid metaheuristic that uses real-time data, stored in building automation systems (e.g., gas/electricity consumption, weather, and occupancy), to find optimal setpoints at the building level and controls the setpoints accordingly. The algorithm consists of metaheuristic (k-nearest neighbor stochastic hill climbing), machine learning (regression decision tree), and self-tuning (recursive brute-force search) components. The control policy uses smart selection of daily setpoints as the control basis, making the control schema complementary to legacy building management systems. To evaluate, we used the DOE reference small office building in all U.S. climate zones and simulated different control policies using EnergyPlus. The proposed algorithm resulted in 31.17% energy savings compared to the baseline operations (22.5 °C and 3 K). The algorithm has a superior performance in all climate zones for the goodness of measure (i.e., normalized root mean square error) with a value of 0.047.

Personal comfort systems (PCS) have potential to fulfill building occupants' personal thermal comfort preferences with great efficiency. But to integrate them into building conditioning, there must be a broader selection of PCS devices available. Design guidance and standards are needed to assure that such devices provide high levels of comfort effectiveness and energy efficiency. This study addresses these needs. A suite of minimum-power PCS devices was built that target body parts significant to alliesthesia—a heated shoe insole, heated/cooled wristpad, small deskfan, and heated/cooled chair. They were tested in a climate chamber under cool and warm conditions using both thermal-manikin and human-subjects. Their efficiency at physically heating/cooling the body is high; the combined suite has a coefficient of performance (COP) of 3.6 for cooling and 0.88 for heating. The subjects' whole-body thermal acceptance and thermal comfort perception were improved by the devices in an additive manner; using the combined suite over 80% of people accepted ambient temperatures of 18°C and 29°C. The PCS ‘corrects’ the ambient temperature towards thermal neutrality by as much as 6.5 K cooling and 3.6 K heating, overcoming building occupants' typical interpersonal thermal differences and making possible large HVAC energy savings in buildings. The idea of temperature corrective power can be the basis of standards for PCS.

Providing personalized energy-use information to individual occupants enables the adoption of energy-aware behaviors in commercial buildings. However, the implementation of individualized feedback still remains challenging due to the difficulties in collecting personalized data, tracking personal behaviors, and delivering personalized tailored information to individual occupants. Nowadays, the Internet of Things (IoT) technologies are used in a variety of applications including real-time monitoring, control, and decision-making due to the flexibility of these technologies for fusing different data streams. In this paper, we propose a novel IoT-based smartphone energy assistant (iSEA) framework which prompts energy-aware behaviors in commercial buildings. iSEA tracks individual occupants through tracking their smartphones, uses a deep learning approach to identify their energy usage, and delivers personalized tailored feedback to impact their usage. iSEA particularly uses an energy-use efficiency index (EEI) to understand behaviors and categorize them into efficient and inefficient behaviors. The iSEA architecture includes four layers: physical, cloud, service, and communication. The results of implementing iSEA in a commercial building with ten occupants over a twelve-week duration demonstrate the validity of this approach in enhancing individualized energy-use behaviors. An average of 34% energy savings was measured by tracking occupants’ EEI by the end of the experimental period. In addition, the results demonstrate that commercial building occupants often ignore controlling over lighting systems at their departure events that leads to wasting energy during non-working hours. By utilizing the existing IoT devices in commercial buildings, iSEA significantly contributes to support research efforts into sensing and enhancing energy-aware behaviors at minimal costs.

This paper provides a systematic approach for quantifying the influence of building size, construction category, climate, occupancy schedule, setpoint, and deadband on HVAC energy consumption in office buildings. Simulating the DOE reference office buildings of three sizes and three construction categories in all United States climate zones, using the EnergyPlus, we conducted several N-way ANOVA analyses to study the interrelationships between setpoints, deadbands and several building related and environment related factors. In summary, daily optimal deadband selection of 0, 1, 2, 4, 5, and 6 K would result in an average energy savings of -70.0, -34.9, -13.7, 9.6, 16.4, and 21.2 %, respectively, compared to baseline deadline of 3 K. Selecting the daily optimal setpoint in the range of 22.5 ± 1 °C, 22.5 ± 2 °C, and 22.5 ± 3 °C would result in an average savings of 7.5, 12.7, and 16.4 %, respectively, compared to the baseline setpoint of 22.5 °C. Additionally, we found that when the outdoor temperature is within -20 to 30°C, the optimal setpoint depends on the building size. We also observed a range of outdoor temperatures (e.g., 9 – 14 °C for small buildings and 8 – 11 °C for medium buildings) where the setpoint selection would only slightly influence the energy consumption. However, the choice of setpoints becomes very influential (up to 30% of energy savings) where the outdoor temperatures are slightly outside the mentioned ranges on either direction. The potential savings from selecting daily optimal setpoints in the range of 22.5 ± 3 °C in different climates and for small, medium and large office buildings, would lead to 10.09 – 37.03%, 11.43 – 21.01%, and 6.78 – 11.34 % savings, respectively, depending on the climate.