CAREER: Human-Centric Automation in the Built Environment

建筑物、电网和建筑环境的其他方面正变得越来越自动化，以满足复杂的社会目标和个人需求。作为一个示例，公用事业需求响应程序目前增加建筑物冷却温度设定点以减少夏季高峰能量负载，但是这可能导致不舒适和沮丧的居住者，他们超越那些设定点以保持舒适。这种覆盖增加了电网的压力，并导致因未能提供承诺的能源管理而造成的财务损失。这个教师早期职业发展计划（CAREER）项目的目标是使建筑环境中的自动化设计能够模拟人类对不断变化的环境温度条件的生理和行为反应，以满足能源管理和居住者舒适度的竞争目标。该项目将使用基于实验室的人类受试者研究和涉及100个消费者家庭的实地研究，以开发方法和模型，了解人类热舒适性和行为（热设定点控制覆盖）如何适应动态热环境。我们的目标是开发一个自适应控制器，采用动态模型的乘员舒适度，这将允许暂时的舒适度偏差，但避免设定点覆盖。该项目将推进美国国家科学基金会的使命，以促进科学的进步和推进国家的健康，繁荣和福利，通过推进复杂的动态，通过联合优化建筑供暖和空调系统的能源消耗和居住者的舒适度的基本理解。与城市职业高中的协同教育计划将把研究成果转化为应用课程。其目标是培训下一代建筑技术人员和工程师，使其在建筑自动化系统的操作中有效和公平。这个CAREER项目的目标是研究人类的热舒适度以及与智能恒温器设备的交互，通过设计避免用户超控的自动控制来缓解能源管理系统与建筑物占用者之间的冲突。该项目的研究计划遵循一个迭代的设计过程，涉及人类心理生理测试，行为建模和控制系统设计。人类行为对环境热瞬变的实验室研究将指导与领先的智能恒温器制造商合作在100个家庭中进行的精简现场实验。人类的心理生理行为动力学模型将开发的基础上产生的实验数据。机器学习技术将用于提高估计器性能。从用户和各种现有的和新颖的鲁棒和最优控制器的相互作用所产生的紧急行为将通过模拟和现场测试进行评估。将考虑到不同人口和环境背景下的公平问题。居住者行为模型和环境控制器设计的开发将被整合到学习练习中，未来的工程师和技术人员将在这些练习中探索建筑环境中以人为本的自动化。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

Buildings, electric grids, and other aspects of the built environment are becoming increasingly automated to satisfy both complex societal objectives and individuals’ needs. As one example, utility demand response programs currently increase building cooling temperature setpoints to reduce peak summer energy loads, but this can lead to uncomfortable and frustrated occupants who override those setpoints to maintain comfort. Such overrides increase strain on the electric grid and induce financial losses resulting from the failure to deliver promised energy management. The objective of this Faculty Early Career Development Program (CAREER) project is to enable the design of automation in the built environment that models human physiological and behavioral responses to changing environmental temperature conditions to satisfy competing objectives of energy management and occupant comfort. This project will use laboratory-based human subject studies and field studies involving 100 consumer households to develop methods and models for understanding how human thermal comfort and behavior (thermal setpoint control overrides) adapt to dynamic thermal environments. The goal is to develop an adaptive controller that incorporates a dynamic model of occupant comfort, which will allow temporary comfort deviations that nevertheless avoid setpoint overrides. The project will advance the NSF mission to promote the progress of science and to advance national health, prosperity, and welfare by advancing a fundamental understanding of the complex dynamics that arise through the joint optimization of building heating and air conditioning system energy consumption and occupant comfort. A synergistic education program with an urban vocational high school will translate research findings into an applied curriculum. The goal of which is to train the next generation of building technicians and engineers to be effective and equitable in the operation of building automation systems. The goal of this CAREER project is to investigate human thermal comfort and interactions with the smart thermostat devices to mitigate conflict between energy management systems and building occupants by designing automated controls that avoid user overrides. This project’s research plan follows an iterative design procedure involving human psychophysiological testing, behavior modeling, and control system design. Laboratory studies of human behavior in response to environmental thermal transients will guide streamlined field experiments to be performed in 100 homes in partnership with a leading smart-thermostat manufacturer. A model of human psychophysiological behavior dynamics will be developed based on the resulting experimental data. Machine learning techniques will be used to improve estimator performance. Emergent behaviors arising from the interaction of users and a variety of existing and novel robust- and optimal-controllers will be evaluated through simulations and field tests. Consideration will be given to issues of equity across populations and environmental contexts. The development of occupant-behavior models and environmental controller design will be integrated into learning exercises in which future engineers and technicians explore human-centric automation in the built environment.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.