Human-Aware Environments for Distributed Building Systems Operations

点击翻译按钮获取中文摘要

Innovative management of urban infrastructure operations has a critical role in enhancing their serviceability, sustainability, and flexibility in integration of dynamic sources of energy. Buildings, a major component of the urban infrastructure, account for majority of energy consumption. This project investigates novel approaches for human-centered and distributed quantification of thermal demand in buildings for efficient thermal energy management. Conventional monitoring and control approaches in buildings are constrained with limited feedback from an environment as well as conservative assumptions that do not reflect the dynamics of their users. To address these limitations, this project investigates a closer system integration between human body thermal response and control mechanisms of building systems. Accordingly, the project investigates non-intrusive sensing and inference methodologies that enable a smart environment to infer the need of its occupants by considering the trade-off between instrumentation, feasibility, and performance. Realization of the research will provide the ground in identifying adaptation potentials at building level and building networks at grid level to improve serviceability of building infrastructure. By leveraging the cross-disciplinary nature of the research, this project integrates findings in undergraduate and graduate courses. It also contributes to the programs that encourage K-12 and undergraduate learners from groups underrepresented in engineering to pursue degrees in STEM fields in collaboration with Virginia Tech CEED center. Human body responses to ambient conditions, which are reflected in cardiopulmonary adjustments, act as a transmitter in an environment. This characteristic is leveraged in this project to shift from human-centric instrumentation for personalized thermal comfort assessment to space-centric by a novel application of Doppler radar systems that act as the sole transceivers. The fundamental requirements of feasibility and performance for a system integration will be investigated through experimental and field validation studies, mathematical modeling, and design of an alternative control framework. Statistical inference techniques, coupled with specialized signal processing frameworks will be developed to enable the application of human physiological response as sensor proxies for distributed feedback in control of building systems? operation. To this end, bio-signal feature engineering and sensitivity analyses will be carried out to develop efficient probabilistic models of thermal response feedback that account for environmental noise interference. Parametric and non-parametric techniques for modeling thermal response of human body will be also utilized to investigate new dimensions to standard thermal sensation metrics. Alternative control scenarios that combine real-time personalized feedback with existing control logic in building systems will be evaluated to (1) assess the efficacy of the control feedback from physiological responses and (2) identify potential energy efficiency improvements in buildings as the main research hypotheses.