CPS: Breakthrough: Low-cost Continuous Virtual Energy Audits in Cyber-Physical Building Envelope

CPS：突破：网络物理建筑围护结构中的低成本连续虚拟能源审计

• 批准号: 1544687
• 负责人: Nirmalya Roy
• 金额: $ 49.81万
• 依托单位: University of Maryland Baltimore County
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-15 至 2020-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1544687&HistoricalAwards=false
• 关键词: CPS; Breakthrough; Low; cost; Continuous

Electricity usage of buildings (including offices, malls and residential apartments) represents a significant portion of a nation's energy expenditure and carbon footprint. Buildings are estimated to consume 72% of the total electricity production in the US. Unfortunately, however, 30% of this energy consumption is wasted. Virtual energy assessment is an approach that can optimize building energy efficiency and minimize waste at a low cost with minimal expert intervention. A virtual energy audit includes a thorough and near real time analysis of different sources of building energy usage, individualized energy footprints of load appliances and devices, and proactive identification of energy holes and air leakages. This project builds a low cost solution that combines the use of non-intrusive single point energy monitoring and low cost IR cameras to provide continuous energy audits. The system will provide continuous virtual audit reports to the landlords or residential users. The system will be deployed in low-income neighborhoods in Baltimore City, Maryland, where poor insulation problems are assumed to be fiscally insurmountable and low cost solutions to determining these issues is important for the landlords. To develop a scalable low cost virtual energy auditing system, this breakthrough research pursues the interfaces of smart building sensing, computing and actuation. The project will be executed under three main research thrust areas. First, it utilizes an autonomous discovery, profiling and rule-based predictive model to capture the relationship between disaggregated power measures and a device's actual usage patterns to pinpoint any abnormal consumption. Second, the PIs develop zero-energy far-infrared imaging sensors for low cost low frequency heat map scanning and air leakage detection. Third, the project engineers and evaluates cyber-physical building sensing system with a control level design perspective for virtual energy auditing that drives the realization of deep energy savings and building efficiency. Additionally, the PIs with collaboration from Constellation will host building energy education projects and workshop where undergraduate, high school, and underrepresented group of students would understand how to design and program energy meters and smart plugs.

建筑物（包括办公室、商场和住宅公寓）的用电量占一个国家能源支出和碳足迹的很大一部分。据估计，建筑物消耗的电力占美国总发电量的 72%。然而不幸的是，其中 30% 的能源消耗被浪费了。虚拟能源评估是一种可以在专家干预最少的情况下以低成本优化建筑能源效率并最大限度减少浪费的方法。虚拟能源审计包括对建筑能源使用的不同来源、负载电器和设备的个性化能源足迹进行全面且近乎实时的分析，以及主动识别能源漏洞和空气泄漏。该项目构建了一个低成本解决方案，结合使用非侵入式单点能源监控和低成本红外摄像机来提供连续的能源审计。该系统将为房东或住宅用户提供连续的虚拟审计报告。该系统将部署在马里兰州巴尔的摩市的低收入社区，那里的绝缘不良问题被认为在财政上是无法克服的，而解决这些问题的低成本解决方案对房东来说非常重要。为了开发可扩展的低成本虚拟能源审计系统，这项突破性研究追求智能建筑传感、计算和驱动的接口。该项目将在三个主要研究领域进行。首先，它利用自主发现、分析和基于规则的预测模型来捕获分类功率测量与设备实际使用模式之间的关系，以查明任何异常消耗。其次，PI开发了零能量远红外成像传感器，用于低成本低频热图扫描和空气泄漏检测。第三，该项目从控制级设计角度对网络物理建筑传感系统进行设计和评估，以进行虚拟能源审计，推动实现深度节能和建筑效率。此外，与 Constellation 合作的 PI 将举办建筑能源教育项目和研讨会，让本科生、高中和少数群体的学生了解如何设计和编程电能表和智能插头。

项目成果

Forecasting Gas Usage for Big Buildings Using Generalized Additive Models and Deep Learning

使用广义加法模型和深度学习预测大型建筑的天然气使用量

• DOI: 10.1109/smartcomp.2018.00092
• 发表时间: 2018-06
• 期刊: 2018 IEEE International Conference on Smart Computing (SMARTCOMP
• 作者: Pathak, Nilavra;Ba, Amadou;Ploennigs, Joern;Roy, Nirmalya
• 通讯作者: Roy, Nirmalya

Longitudinal energy waste detection with visualization

可视化纵向能源浪费检测

• DOI: 10.1145/3137133.3137162
• 发表时间: 2017-01
• 期刊: BuildSys '17 Proceedings of the 4th ACM International Conference on Systems for Energy-Efficient Built Environments
• 作者: Lachut, David;Pathak, Nilavra;Banerjee, Nilanjan;Roy, Nirmalya;Robucci, Ryan
• 通讯作者: Robucci, Ryan

RAM: Radar-based activity monitor

RAM：基于雷达的活动监视器

• DOI: 10.1109/infocom.2016.7524361
• 发表时间: 2016-04-10
• 期刊: IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications
• 作者: Md Abdullah Al Hafiz Khan;Ruthvik Kukkapalli;Piyush Waradp;e;e;Sekar Kul;aivel;aivel;Nilanjan Banerjee;Nirmalya Roy
• 通讯作者: Nirmalya Roy

Iterative signal separation assisted energy disaggregation

迭代信号分离辅助能量分解

• DOI: 10.1109/igcc.2015.7393701
• 发表时间: 2015-12-14
• 期刊: 2015 Sixth International Green and Sustainable Computing Conference (IGSC)
• 作者: Nilavra Pathak;Nirmalya Roy;A. Biswas
• 通讯作者: A. Biswas

Wearable Sensor-Based Location-Specific Occupancy Detection in Smart Environments

智能环境中基于可穿戴传感器的特定位置占用检测

• DOI: 10.1155/2018/4570182
• 发表时间: 2018-04-01
• 期刊: Mob. Inf. Syst.
• 作者: Md Abdullah Al Hafiz Khan;Nirmalya Roy;H. M. S. Hossain
• 通讯作者: H. M. S. Hossain