pv magazine Outdoor Test: Summary of 3 Years of Operation
By Huatian Xu, George Touloupas, and Ryan Li
This article was originally published in pv magazine – May 2021 edition. Learn more about the pv magazine test here.
The pv magazine test array in Xi’an, China, is being dismantled and upgraded this month. CEA Director of Technology and Quality, George Touloupas, Senior Technology and Quality Manager, Huatian Xu, and Ryan Li, Junior Engineer for Technology and Quality, have taken the opportunity to analyze three years of performance data, in addition to running modules from the array through another series of lab measurements.
History and Upgrade of Test Field
The first 13 samples were installed in the pv magazine outdoor PV module test field at the rooftop of our partner Gsola’s facility in Xi’an, China, back in May 2018, and the first results were published in the September 2018 issue of pv magazine. In total, 23 different samples from various manufacturers have been installed in the last 3 years. Some of these samples have been removed from the array for various reasons.
Recently we have seen a proliferation of new module designs, form factors and electrical specifications as the introduction of new wafer sizes (166 mm, 182 mm, 210 mm) is becoming common. In order to introduce these new module designs in the outdoor test field we had to upgrade it, so that its electrical and mechanical hardware is compatible.
The major upgrades were:
The mounting structure is being rebuilt to accommodate the larger dimensions of 600W + modules.
As the new designs have much higher powers, we are replacing the existing optimizers and installing new, high power microinverters, provided by AP System, that can support higher currents and powers as well as accommodate the very big differences of electrical characteristics of the new generation of modules.
To implement these upgrades, all samples were uninstalled, so no field data will be available for this month. We therefore decided to take this opportunity to conduct EL and power measurement tests of the samples and analyze the historical production data and attempt to draw some conclusions on the degradation rates and the relationship of the initial characterization metrics of Low Irradiance Loss and Pmax Temperature Coefficient to the performance ratios of the samples.
Degradation
As the duration of operation of these modules ranges from 5 to 33 months, it makes sense to compare the average monthly degradation values, with measurement uncertainty of 1% depicted by the error bars in the graph. The warrantied degradation is taken by dividing the yearly degradation stated in the PV module datasheets by 12 (months in a year). All modules have been light soaked, so the there is no initial light induced degradation in the degradation figures.
Performance Ratio
We chose the Performance Ratio (PR) metric to analyze the generation data of each products. Figure 2 below shows the overall distribution of daily PR calculated from historical data, from which we can observe an almost normal distribution ranging from around 50% to 130%.
The abnormally low or high PR values may have been caused by electricity meter related issues during the 3 years of operation, or partial shading in some circumstances. Therefore, for further analysis, the data with performance ratio lower than 30% and higher than 200% were dismissed as outliers, with the remaining of the daily data used for further analysis.
Before being installed in the outdoor field, the samples were subject to various tests, including measuring the Pmax (maximum power) temperature coefficient, which influences performance at high temperatures, and measuring the Low Irradiance Loss, which influences performance in low light conditions.
The chart in Figure 3 plots all the test modules to their low irradiance loss and Pmax temperature coefficient and also marks the PR of each sample. Best to worse performers are colored from green to red.
As expected, the overall performance of bifacial modules is outstanding compared to monofacial modules.
The Pmax temperature coefficient shows a good correlation to PR, as the green PR values are concentrated in the right-hand area of the chart, within the green rectangle, that has lower absolute temperature coefficient values.
The low irradiance loss performance does not correlate with PR. This is probably because data points from low irradiance periods are less predominant in the overall data.
Figure 4 shows the performance ratio of all the tested monofacial modules on high temperature days.
Temperature and PR
The influence of temperature on PR is significant, as shown in Figure 4.
Apart from some data points with pronounced irregularities on the right-hand side of the chart, the average reduction on PR is between 0.84% - 1.06% per degree Celsius for the monofacial modules. At a fixed temperature, we can see the trends, shown as dotted lines in Figure 4. The modules with smaller absolute Pmax temperature coefficients have higher PR. As expected, the Pmax temperature coefficient of PV modules influences their performance.
In contrast to the Pmax temperature coefficient, the relationship between low irradiance loss and PR seems to be less consistent between different products (see Figure 5 above). As expected, the trendlines show that modules with higher low irradiance losses will have lower PR. Modules appear to have better PR on days with lower irradiance, which may be due to lower operating temperatures.
Future Plans of Outdoor PV Test Field
The upgrade of the outdoor test field is expected to be completed in April. Several older products installed in early 2018 will be taken out to make room for the installation of new generation modules, on which we expect to gather soon enough data to further our analysis of their performance.
George Touloupas is CEA’s Director, Technology and Quality; Huatian Xu, Senior Manager, Technology and Quality at CEA; Ryan Li, Junior Engineer, Technology and Quality at CEA
Test Cooperation
pv magazine test is a cooperative effort involving pv magazine, CEA, and Gsolar. All testing procedures are carried out at Gsolar’s test laboratory in Xi’an, China. CEA supervises these tests and designed both the indoor and outdoor testing procedures.
