Title:	PV Fault Detection Threshold at the Module, String, and Inverter Levels
Author(s):	M. Matam, H. Seigneur
Keywords:	Monitoring, I-V Curve, Time Series, Photovoltaic (PV), PV Plant, Machine Learning
Topic:	PV Systems – Modelling, Design, Operation and Performance
Subtopic:	Operation, Performance and Maintenance of PV Systems
Event:	38th European Photovoltaic Solar Energy Conference and Exhibition
Session:	5CV.2.49
Pages:	1284 - 1289
ISBN:	3-936338-78-7
Paper DOI:	10.4229/EUPVSEC20212021-5CV.2.49
Price:	0,00 EUR
Document(s):	paper, poster

Abstract/Summary:

---

This paper presents the low and high granular data significance in detecting and classifying the faults in a solar PV fault. In this experimental study, the low granular inverter measurements were collected every 5s and the high granular string and module I-V curves were collected every 40 min. Mainly top four PV faults were induced on the PV modules to understand their impact and signatures they leave at all the three measurement stages. Finally, statistical Analysis of Variance (ANOVA) p-values were computed with the baseline measurements. Besides, box plot analysis has been performed. The results reveal that strings data is sufficient to detect the faults, thus, no need to measure, acquire, and store a few tens of modules data.