Predictive Data Analytics for Energy Incident Management (EIM) in Smart Buildings: A Data Science Approach

Amid escalating global energy demand and the imperatives of enhanced energy efficiency, the incorporation of Smart Grids (SG) has emerged as a necessity. Buildings, a significant contributor to energy consumption, must evolve towards becoming “smart” through the deployment of appropriate Energy Management Systems (EMS). Central to any EMS is Energy Incident Management (EIM), an essential tool in the timely detection and alerting of potential energy incidents and faults. This paper delves deeper into the complexities of designing a real-time, energy-efficient EIM, which extracts and processes data while leveraging Machine Learning (ML) predictive analytics to forecast possible energy incidents. Our methodology harnesses principles of Data Science, whereby data from diverse sources is amassed, purified, and readied for use by the ML predictive analytics model. To implement our EIM, we embrace a Cloud-based strategy, taking advantage of existing Cloud technologies to encapsulate all phases of the Data Science process within a comprehensive, unified framework (IDE).