文章基本信息

标题：Real-Time Nuisance Fault Detection in Photovoltaic Generation Systems Using a Fine Tree Classifier
本地全文：下载
作者：Collin Barker ; Sam Cipkar ; Tyler Lavigne 等
期刊名称：Sustainability
印刷版ISSN：2071-1050
出版年度：2021
卷号：13
期号：4
页码：2235
DOI：10.3390/su13042235
语种：English
出版社：MDPI, Open Access Journal
摘要：Nuisance faults are caused by weather events, which result in solar farms being disconnected from the electricity grid. This results in long stretches of downtime for troubleshooting as data are mined manually for possible fault causes, and consequently, cost thousands of dollars in lost revenue and maintenance. This paper proposes a novel fault detection technique to identify nuisance faults in solar farms. To initialize the design process, a weather model and solar farm model are designed to generate both training and testing data. Through an iterative design process, a fine tree model with a classification accuracy of 96.7% is developed. The proposed model is successfully implemented and tested in real-time through a server and web interface. The testbed is capable of streaming in data from a separate source, which emulates a supervisory control and data acquisition (SCADA) or weather station, then classifies the data in real-time and displays the output on another computer (which imitates an operator control room).