Intelligent single-point energy consumption monitoring for intensive energy sectors applications

智能电表被认为是智能电网演进中部署自动计量基础设施(AMI)的关键组件。尽管这些电表能够向电力公司和用户报告总能耗以及使用时间(TOU)，但在工业设施、商业建筑、大学校园等能源密集型行业，仍然缺乏有效的流程和方法来最大限度地减少能源浪费。这些大型行业的特点是设备滞留，这使得从经济和实用角度监测单个设备层面的电能消耗是不可行的。鉴于这一事实，确定特定设备中的能源浪费、选择用于需求侧供应的候选负荷、设备故障排除以及在考虑TOU的情况下为该设备找到最佳运行策略，同时确保不损害整个行业的目标/目标的过程变得具有挑战性。需要大量的研究工作来实现非侵入性且经济高效的能源消耗监测方法，为实现更智能、高效和准确的计量基础设施铺平道路。拟议的研究工作旨在通过开发基于单点传感的AMI智能算法和创新的无监督学习来细分目标行业的聚合能源消耗，从而更好地量化用电量和识别能源浪费，从而弥合现有能源监测系统与智能能源消费之间的差距。通过与该项目的行业合作伙伴的密切合作，拟议的研究工作将为海德鲁一号提供必要的信息，这些信息将有助于减少对其输电/配电系统的影响，延长其资产的使用寿命/性能，扩大西门子加拿大公司能源监测产品的能力，同时确保低产品成本和更多的能源消耗信息可获得性。此外，拟议的研究工作还将有助于安大略省发电公司降低发电生产成本，减少碳足迹。

Smart meters are considered key component in the deployment of automatic metering infrastructure (AMI) in the smart grid evolution. Despite such meters' capabilities in reporting the total energy consumption along with time of use (TOU) to electric utilities and consumers, there is still lack of effective processes and methodologies to minimize the energy waste in intensive energy sectors such as industrial facilities, commercial buildings, university campuses, etc. Such large sectors are characterized by stranded equipment, which makes monitoring the electric energy consumption at individual equipment level from economic and practical perspectives infeasible. Given this fact, the process of identifying the energy waste in specific equipment, selecting candidate loads for demand side provision, equipment troubleshooting, and finding optimal operating strategies for this equipment considering TOU while ensuring not to compromise the whole sector goals/objectives become challenging to realize. Substantial research effort is needed to achieve a non-intrusive and cost-effective energy consumption monitoring approach that paves the road toward realizing smarter, efficient, and accurate metering infrastructure. The proposed research work aims to bridge the gap between existing energy monitoring system and smart energy consumption through the development of intelligent algorithms for AMI based on single-point sensing and innovative un-supervised learning to breakdown the aggregated energy consumption in the targeted sectors hence better quantify the electricity usage and identify the energy waste. With close co-operation with the industrial partners of this project, the proposed research work will supply Hydro One with the necessary information that will help reducing the impact on their transmission/distribution systems, increase the lifetime/performance of their assets, expand the capabilities of the energy monitoring products of Siemens Canada while ensuring low product cost and more energy consumption information accessibility. Moreover, the proposed research work will also help Ontario Power Generation to reduce their generation production cost and reduce their carbon foot print.