Experience With a Physics-Based Anomaly Detection Algorithm to Detect DC Faults at Utility PV Plants

String, combiner, and tracker faults comprise a significant reduction in PV plant capacity. A method is presented to detect these faults using data typically monitored at utility-scale plants. The detection algorithm has been validated using aerial infrared data and the accuracy is presented across...

Full description

Saved in:
Bibliographic Details
Main Authors: Sheppard, Scott, Koskey, Steven, Teasley, Corson, Perullo, Christopher, Fregosi, Daniel, Li, Wayne
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Anomaly detection
Data models
Detection algorithms
Manuals
Meteorology
Monitoring
Photovoltaic systems
Satellites
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:String, combiner, and tracker faults comprise a significant reduction in PV plant capacity. A method is presented to detect these faults using data typically monitored at utility-scale plants. The detection algorithm has been validated using aerial infrared data and the accuracy is presented across a range of sites, conditions, and scenarios, including automated versus manual model configuration and satellite versus ground-based weather data.
ISSN:2995-1755
DOI:10.1109/PVSC57443.2024.10749459