A Data-driven Situational Awareness Approach to Monitoring Campus-wide Power Consumption (original) (raw)

The emerging of big data and data analytics has been an enabler to address key technical challenges in the energy sector. Situational awareness capabilities such as real-time sensor monitoring, anomaly detection, and timely anomalies warning are imperative to the facility operator of large scale energy distributing network. As a part of Georgia Tech’s Smart Campus Initiative, substantial volume and varieties of data obtained from multiple layers of sensor networks throughout the Georgia Tech campus are used to support situational awareness through real-time operations monitoring and energy usage tracking and assessment. More specifically, in this paper, a proposed data-driven approach is presented to model, simulate, and forecast the short-term electricity demand by factoring in available time information and weather data. The predictive results from the model are compared with the online readings collected from electricity meters of interest deployed onsite. As such, once an anomaly is detected and diagnosed with a sufficient confidence level, a corresponding warning can be issued to appropriate facilities personnel in a prompt manner, increasing the possibility to prevent potential disruption to the day-to-day campus operation.

Sign up for access to the world's latest research.

checkGet notified about relevant papers

checkSave papers to use in your research

checkJoin the discussion with peers

checkTrack your impact