Human-Centric Sustainable Spaces
Learning Personal Thermal Comfort & Human in the Loop Control of HVAC Systems
Buildings are one of the largest consumers of natural resources. 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 make the prediction of thermal comfort more difficult.
Thus, it requires to be monitored over time. In this research project, we develop an online learning approach for modeling and quantifying personalized thermal comfort via a dynamic Bayesian network. We develop a novel infrared thermography based technique to monitor an individual’s thermoregulation performance and thermal comfort by measuring the skin temperature on several points on human face, which has a high density of blood vessels and is not covered by clothing. Unlike other models, our method requires no continuous user input or interaction. We also develop a model-free control policy for optimizing HVAC energy consumption that allows the integration of occupants’ real-time thermal comfort. The control policy is an adaptive hybrid metaheuristic that takes in real-time data from the building automation systems to select optimal setpoints. Our results show that the implementation of the algorithm results in 29.77 % energy savings compared to baseline operations.
Activity-Driven & User-Centered Automation of Appliances
We investigate an activity-based approach to improve energy efficiency in buildings by providing the occupants with both a detailed energy consumption feedback and also an automation system that controls the operation of appliances and lighting systems in buildings. Depending on the occupant preferences, our proposed automation system can fully or partially control the appliances and lighting systems, based on a set of generated rules that are static or dynamic. In line with our objectives in this study, we propose and teste a hybrid activity recognition algorithm
that uses machine learning and ontological reasoning to detect activities and the associated personalized energy consumption down to appliance level. To identify the potentials in improving energy efficiency, we explore a waste detection framework that recognizes the wasted energy consumption associated with occupant activities. Based on the outcome from our proposed activity recognition and waste energy detection algorithms, and also automation preferences, which can change both by context and time, a set of automation rules to control appliances and lighting system is generated, using preference-based hierartical task network planning. To formulate automation preferences based on occupant characteristics and different activity-related automation contexts, we use adjustable autonomy, in which varying levels of automation (full automation, adaptive automation, inquisitive automation and no automation) are applied to control appliances and lighting systems.
Building Occupancy Modeling & Occupancy-Load Relationships for Energy Efficiency
with building heating/cooling controls is a complex problem as occupancy is stochastic in nature, and there exists heat transfer among zones of a building, as well as heat gain and loss through a building’s envelope. Since there is no systematic understanding of the relationships between occupancy and loads to achieve energy efficiency, this project has also focused on the types and ways of modeling the relationships that significantly influence HVAC system energy efficiency. Specifically, two relationships have been identified from the two perspectives of occupancy transition, which represents the switch between real-time occupied and unoccupied statuses and occupancy diversity which represents the difference in long-term occupancy. As a more feasible and reliable venue for carrying the investigations, this project has introduced a multi-level calibration framework to calibrate building energy model at multiple levels simultaneously.
Occupancy is one of the most important factors impacting actual demands for HVAC systems. However designed occupancy in commercial buildings rarely represents actual occupancy. In order to leverage HVAC energy efficiency based on occupancy, this project has focused on non-intrusive building occupancy modeling including real-time occupancy (time-sequenced occupancy status changes) and long-term occupancy (typical-weekday/weekend presence/number probability). Ambient sensors are deployed and the relationships between ambient factor and occupancy have been explored to model occupancy. Occupancy awareness is improved for applications at the building level, by which the model is trained in one space and used in other geometrically similar spaces. Integrating occupancy
Building Level Energy Management System (BLEMS)
The importance of efficiency in building energy consumption has assumed great urgency due to fast depleting energy resources and increasing environmental pollution. To address this urgent and practical issue, the BLEMS project aims to develop solutions for bringing together ad-hoc legacy. Energy Management System (EMS) configurations under a single unified framework that makes them interoperable. Communication networks are implemented to transmit building sensor data to a centralized brain, which in response, sends and controls information to system actuators. In this way, the framework alleviates the apparent limitations of EMS legacy systems, including security, reliability, extensibility, self-management, and self-dispersed optimal energy allocation. By studying the human behavior in regard to energy consumption and understanding the building behavior in regard to device operations, the behavior-driven self-contained BLEMS system serves to maximize user/occupant comfort levels while simultaneously minimizing energy usage and/or energy cost.
A Framework for Enabling Energy-Aware Smart Facilities
The goal of this research is to identify ways to inexpensively provide specific information about energy consumption in buildings and facilitate conservation. Signal processing, machine learning, and data fusion techniques are developed to extract actionable information from whole-building power meters and other available sensors. The objectives are: (1) to create a framework for obtaining disaggregated, appliance-specific feedback about electricity consumption in a building by extracting high-value information from low-cost data sources; and (2) to investigate and develop data mining and machine learning algorithms for making use of appliance-specific electricity data, in order to provide users with recommendations on how to optimize their energy consumption and understand the effects of their energy-related decisions. The scientific merit of the project is the development of a framework for evaluating energy-use-disaggregation methods according to their value for promoting energy conservation. The resulting data sets are large enough to produce significant conclusions about the feasibility and effectiveness of the technology, and allow for the development of new models about the trends and patterns of appliance usage in buildings.