CAREER: Urban Informatics for Smart, Sustainable Cities: Toward a Data-Driven Understanding of Metropolitan Energy Dynamics

CBET 1653772; PI：Kontokosta，Constantine E.该项目将使用数据驱动的方法，通过城市能源需求、行为和基础设施的耦合模型，推进对城市能源动态的基本理解。这项工作旨在通过支持基于证据的城市能源政策，公共和私人决策以及基础设施投资来最大限度地发挥影响，以更有效地设计，实施和评估能源减少战略。该研究利用PI获得的多样化、全面且独特的数据集，旨在解决以下问题：（1）是什么驱动城市内部和城市之间的能源使用？(2)建筑能耗的社会技术动态是什么？(3)节能措施和改造机会如何以及为什么会因建筑类型、城市和地区而异？(4)城市能源使用的时空模式是什么？(5)随着时间的推移，城市和区域政策如何影响能源效率和成本节约？城市信息学和大都市能源动力学的工作重点是开发新的分析方法，再加上来自美国和国际城市的一系列建筑，土地使用和能源数据，以促进对城市能源需求和温室气体排放的模式和决定因素的基本理解。这将通过整合土木和系统工程、数据科学和计算社会科学的方法来实现，以开发数据驱动的模型，通过从大数据中提取可操作的情报来支持决策。这项研究将（1）整合数万座建筑物和多个城市的建筑物，社区和城市层面的数据，（2）利用城市能源数据的新来源来创建一个大型的，非自选的数据集，（3）同时检查物理，环境，社会，和城市动态的行为组成部分，以创建跨大都市地区的能源需求和减少潜力的多标量模型。这项研究力求提供分析的严谨性，以支持客观的、以证据为基础的政策，这些政策将建立一个框架，对拟议和执行的战略进行以业绩为导向的评价。该教育计划将促进发展一个在建筑能效、城市信息学和城市可持续性方面接受培训的学生网络，并提高公众对应对城市能源挑战的规模和重要性的认识。研究和教育活动将建立在与城市机构，行业合作者和MetroLab Network（由34个城市-大学合作伙伴组成，专注于城市挑战的数据解决方案）的现有关系基础上，为美国和全球的智能，可持续城市提供基础。

CBET 1653772; PI: Kontokosta, Constantine E. This project will use data-driven methodologies to advance fundamental understanding of urban energy dynamics through a coupled model of urban energy demand, behavior, and infrastructure. The work seeks to maximize impact by supporting evidenced-based urban energy policy, public and private decision-making, and infrastructure investment to more effectively design, implement, and evaluate energy reduction strategies. Using a diverse, comprehensive, and unique collection of data acquired by the PI, the research aims to tackle the following: (1) What drives energy use within and across cities? (2) What are the socio-technical dynamics of building energy consumption? (3) How and why do energy conservation measures and retrofit opportunities vary by building type, city, and region? (4) What are the spatial-temporal patterns of energy use in cities? (5) How do urban and regional policies impact energy efficiency and cost savings over time?The work in urban informatics and metropolitan energy dynamics is focused on developing new analytical approaches, coupled with an array of building, land use, and energy data from U.S. and international cities, to advance the fundamental understanding of the patterns and determinants of urban energy demand and GHG emissions from the built environment and their impacts on human well-being. This will be achieved by integrating methods from civil and systems engineering, data science, and computational social science to develop data-driven models to support decision-making through the extraction of actionable intelligence from big data. This research will (1) integrate an array of building, neighborhood, and city level data across tens of thousands of buildings and multiple cities, (2) utilize new sources of urban energy data to create a large, non-self-selected dataset, and (3) simultaneously examine physical, environmental, social, and behavioral components of urban dynamics to create a multi-scalar model of energy demand and reduction potentials across metropolitan areas. The research seeks to provide the analytical rigor to support objective, evidenced-based policies that will create a framework for performance-driven evaluation of proposed and implemented strategies. The education plan will facilitate development of a network of students trained in building energy efficiency, urban informatics, and urban sustainability, as well as foster greater public awareness of the scale and importance of addressing energy challenges in cities. The research and education activities will build on an existing relationship with city agencies, industry collaborators, and the MetroLab Network - a group of 34 city-university partnerships focused on data solutions to urban challenges - to provide the foundation for smart, sustainable cities in the U.S. and globally.