Data Cleaning in Large Customer-Owned Electricity Meter Systems
Electricity meter data is crucial for both utility companies and their customers. Meter data has a variety of uses such as energy and power consumption analysis, grid optimization, infrastructure planning, and achieving environmental goals. Meter Data Management System (MDMS) is a large customer-owned meter network with over 20,000 meters operated by the Department of Defense (DoD). MDMS meters measure energy and demand over fixed time intervals and report this data to a centralized server that stores and compiles data into reports. Users can change the parameters of these reports to suit their needs. Due to its size, complexity, and communication requirements, a variety of systemic MDMS issues resulted in erroneous meter data. Specifically, seven meter data issues were observed, and their causes were analyzed and defined. A novel data cleaning method was developed using a sequential series of seven algorithms to clean the data, creating more accurate meter data. Results are presented for West Point's MDMS data.
Authors
- Nicholas G. Barry (ORCID: https://orcid.org/0000-0002-8474-5282)
- Aaron St. Leger (ORCID: https://orcid.org/0000-0002-1685-2756)
- Bach Nguyen
Institutions
- United States Military Academy (US)
Publication Details
- Published
- 2024-10-13
- DOI
- https://doi.org/10.1109/naps61145.2024.10741871
- Primary Topic
- Electricity Theft Detection Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00