III: Small: Cyber Physical Mappings - Empower Building Analytics at Scale

建筑对人类健康、生产力、舒适度和能源消耗有着深远的影响。例如，建筑运营是美国最大的单一能源消耗，占电力消耗的70%，占总能源消耗的40%。过敏原、噪音水平和阳光的可用性都会影响健康和幸福，特别是考虑到美国人平均90%的时间都在建筑物里度过。热舒适、二氧化碳和污染物浓度等室内条件对人类生产力的影响为8-11%，对国民经济具有重要影响。通过分析引擎，可以从典型建筑物内已经存在的数千个传感和控制点收集和分析数据，从而大大提高国家建筑物的性能。然而，数据本身本身并不具有任何意义，因此必须手动提供有关每个传感器和控制器的上下文（也称为元数据），以便分析引擎能够解释数据。对于单个建筑物，这种昂贵的手动过程可能需要数天或数周的时间，并且是将建筑物分析应用于大量建筑物的主要障碍。该项目创建了一些工具来自动推断数据流的元数据，例如产生数据的传感器或控制器的类型及其与建筑物中其他传感器、设备或房间的关系。该方法基于这样一个假设，即建筑物中的数据是根据天气模式、设备操作模式和在世界各地的许多建筑物中观察到的常见设计模式进行结构化的。元数据推理利用这种结构，基于其他点或其他建筑物的已知元数据，快速轻松地为大量传感和控制点创建新的元数据值。它沿着三个主要研究方向开发了新的基于学习的技术：1)单个点的价值推断，2)传感器之间的关系推断，以及3)从建筑物管理系统的建筑物管理人员交互访问行为中推断潜在的元数据。这项研究使行业和机构能够更容易地将建筑分析应用于新建筑，而无需手工绘制。它在多个指标上对美国建筑的平均性能产生影响，包括人体健康、生产力、舒适度和能耗。此外，拟议的研究包括在数据挖掘和网络物理系统领域开发全新的方法和技术，并将作为开源代码发布。这些研究活动将被纳入学生培训和教育的教材。研究生和本科生的研究人员都将参与这项研究的所有阶段，我们将从代表性不足的群体中招募学生参与这项研究。如果成功，这些技术将推广到其他类型的C活动，如人体健康监测、基础设施监测或智能交通系统，其中结构可以类似地用于帮助推断传感器或控制器的物理环境。

Buildings have profound impact on human health, productivity, comfort, and energy consumption. For example, building operation is the single largest energy consumer in the US, accounting for 70% of electricity consumption and 40% of total energy consumption. Allergens, noise levels, and the availability of sunlight affect health and well-being, especially given that on average Americans spend 90% of their time in buildings. Indoor conditions such as thermal comfort and CO2 and pollutant concentrations have been shown to affect human productivity by 8-11%, which has an important effect on the national economy. The performance of the nation's buildings can be significantly improved with analytics engines that collect and analyze data from the thousands of sensing and control points that already exist within a typical building. However, data alone does not inherently have any meaning, and so a person must manually provide the context (also called metadata) about every sensor and controller so that the analytics engine can interpret the data. This costly manual process can take days or weeks for a single building and is a major obstacle for applying building analytics to a large number of buildings.This project creates tools to automatically infer the metadata of data streams, such as the type of sensor or controller that produced the data and its relation to other sensors, equipment, or rooms in the building. The approach is based on the hypothesis that the data in buildings is structured due to weather patterns, equipment operation patterns, and common design patterns that are observed in many buildings around the world. Metadata inference exploits this structure to quickly and easily create new metadata values for a large number of sensing and control points based on known metadata of other points or other buildings. It develops new learning-based techniques along three main research thrusts: 1) value inference of individual points, 2) relationship inference between sensors, and 3) latent metadata inference from building managers' interactive access behaviors with a building management system. This research enables industry and institutions to more easily apply building analytics to new buildings with minimal or even no manual mapping required. It generates impact on average US building performance along multiple metrics, including human health, productivity, comfort, and energy consumption. In addition, the proposed research includes the development of fundamentally new methods and techniques in the fields of data mining and cyber-physical systems, and they will be released as open-sourced code. The research activities will be incorporated into teaching materials for student training and education. Both graduate and undergraduate researchers will be involved in all phases of this research, and we will engage and recruit students from underrepresented groups to participate in this research. If successful, these techniques will generalize to other types of C activities such as human health monitoring, infrastructure monitoring, or smart transportation systems where structure can similarly be used to help infer the physical context of a sensor or controller.