Profiling Real-Time Electricity Consumption Data for Process Monitoring and Control
- ORNL
Today, smart meters serve as key assets to utilities and their customers because they are capable of recording and communicating real-time energy usage data; thus, enabling better understanding of energy usage patterns. Other potential benefits of smart meters data include the ability to improve customer experience, grid reliability, outage management, and operational efficiency. Despite these tangible benefits, many utilities are inundated by data and remain uncertain about how to extract additional value from these deployed assets outside of billing operations. One way to overcome this challenge is the development of new metrics for classifying utility customers. Traditionally, utilities classified their customers based on their business nature (residential, commercial, and industrial) and/or their total annual consumption. While this classification is useful for some operational functions, it is too limited for designing effective monitoring and control strategies. In this paper, a data mining methodology is proposed for clustering and profiling smart meters data in order to form unique classes of customers exhibiting similar usage patterns. The developed clusters could help utilities in identifying opportunities for achieving some of the benefits of smart meters data.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Laboratory Directed Research and Development (LDRD) Program
- DOE Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1087475
- Resource Relation:
- Conference: Industrial and Systems Engineering Research Conference, San Juan, PR, USA, 20130518, 20130522
- Country of Publication:
- United States
- Language:
- English
Similar Records
AGGREGATE: dAta-driven modelinG preservinG contRollable dEr for outaGe mAnagemenT and rEsiliency (Final Report)
Austin Sustainable and Holistic Integration of Energy Storage and Solar PV [Austin SHINES]. Final Report, Version 2