A methodology to analyze photovoltaic tracker uptime. (17th April 2018)

• Record Type: Journal Article
• Authors: Ruth, Dan; Muller, Matthew
• Is Part Of: Progress in photovoltaics. Volume 26:Number 7(2018)
• Page Start: 491
• Page End: 501
• Publication Date: 2018-04-17
• Subjects: clear‐sky modeling -- photovoltaic performance -- tracking; Solar cells -- Periodicals; Photovoltaic cells -- Periodicals; Solar power plants -- Periodicals; 621.31245
• DOI: 10.1002/pip.3002
• Languages: English
• ISSNs: 1062-7995

Abstract

Abstract: A metric is developed to analyze the daily performance of single‐axis photovoltaic (PV) trackers. The metric relies on comparing correlations between the daily time series of the PV power output and an array of simulated plane‐of‐array irradiances for the given day. Mathematical thresholds and a logic sequence are presented, so the daily tracking metric can be applied in an automated fashion on large‐scale PV systems. The results of applying the metric are visually examined against the time series of the power output data for a large number of days and for various systems. The visual inspection results suggest that overall, the algorithm is accurate in identifying stuck or functioning trackers on clear‐sky days. Visual inspection also shows that there are days that are not classified by the metric where the power output data may be sufficient to identify a stuck tracker. Based on the daily tracking metric, uptime results are calculated for 83 different inverters at 34 PV sites. The mean tracker uptime is calculated at 99% based on 2 different calculation methods. The daily tracking metric clearly has limitations, but as there is no existing metrics in the literature, it provides a valuable tool for flagging stuck trackers. Abstract : A metric is developed to automatically determine if a single‐axis photovoltaic (PV) tracker is following the sun or stuck in a fixed position on any clear‐sky day. The metric is validated and assessed and then used to calculate the uptime of for single‐axis trackers associated with 83 different inverters and 34 PV sites. The average uptime for these systems is calculated to be 99%.