Newer and larger solar plants perform better, while older plants pick up steam with repowering


The US Department of Energy’s Lawrence Berkeley National Laboratory (LBNL) released an analysis of full plant generation degradation of the United State’s 31GW solar power fleet. The analysis finds that total system degradation is averaging 1.2% per year, greater than what is projected in many observed power purchase agreements.

However, the analysis finds two nuances in the data that paint a picture of stronger performance. First – newer (post 2014) and larger (greater than 25MWac) facilities are showing only 0.7% of system wide degradation. The second nuance in the data is that older facilities show a bump in generation in their later years. This suggests that repairs start in year seven, result in increased generation in year eight, and that performance improvements in subsequent years are sustained.

In fact, looking at year nine in the above chart – it appears that older facilities nearly reach their year one capacity factors. If this data pans out, it speaks volumes to repowering opportunities in the future.

These two data points suggest that newer and larger facilities are improving in both quality and operations/maintenance compared to their predecessors. It implies that older facilities are being upgraded or repaired by asset owners seeking to improve returns.

Often, solar assets in the US are resold to new owners after the fifth full year of operation due to the Investment Tax Credit structure. The bump in year eight and nine may be attributed to the asset transfer occurring in year six, as analysis and repairs occur in year seven, and results show up shortly thereafter.

In the analysis, Plant-level performance and degradation of 31 GWDC of utility-scale PV in the United States, the authors attempt to tease out recoverable versus non-recoverable degradation losses.

Recoverable losses are defined as correctable performance issues, like soiling or tracker malfunction, while unrecoverable losses are typically due to solar module delamination or solar cell hot spots. Naturally, these losses are larger than those of solar modules alone, as modern solar panels are typically warrantied at or below 0.5% of degradation per year.

The vertical line in the chart demonstrates how singular events – failures of fuses, trackers, breakers, and inverters – can briefly, but dramatically cut solar power facility outputs. The author notes that inverters or breakers may cut 100% of plant output, and that fuses can cut around half of a plant’s output. Individual tracker failures and curtailment may result in a 20-30% reduction to a plant’s output.

Of course, solar modules degrade slowly, and indefinitely, thus the wide and ever increasing dark blue bar. Interestingly, when the authors modeled losses based on solar module manufacturers, they found no statistically significant differences amongst the standard silicon module manufacturers. However, plants powered by thin film products – namely First Solar – showed 30% less annual degradation.

The report cited a selection of variables that show statistically significant differences in generation:

  • Vintage: Post-2014 plants (-1.1%/year, ±0.3%) have declined less than pre-2015 plants (-1.2%/year, ±0.1%)
  • Capacity: Plants of 25 MWAC or larger (-0.8%/year, ±0.2%) have declined less than plants less than 25 MWAC (-1.4%/year, ±0.1%)
  • DC:AC ratio (ILR): Projects with a DC:AC ratio of 1.25 or greater (-1.1%/year, ±0.2%) degrade less than projects with lower ratios (-1.3%/year, ±0.2%)
  • Tracking: Fixed-tilt plants (-1.2%/year, ±0.2%) have declined less than plants with single-axis tracking (-1.3%/year, ±0.2%)
  • Module tech: Plants using thin-film modules (-1.0%/year, ±0.2%) have declined less than plants using Si modules (-1.3%/year, ±0.1%)
  • COMBINATION: Newer (post-2014), larger (≥25 MWAC) plants with higher DC:AC ratios (≥ 1.25) have declined less (-0.7%/year, ±0.4%) than their counterparts (-1.4%/year, ±0.2%)
  • Solar resource: Plants at sites with an average long-term global horizontal irradiance of less than 210 W/m2 (-1.0%/year, ±0.2%) have declined less than plants at sunnier sites (-1.2%/year, ±0.1%)
  • Site temperature: Plants with an average site temperature of less than 15 degrees Celsius (-0.9%/year, ±0.2%) have declined less than plants at warmer sites (-1.3%/year, ±0.1%).

While the primary purpose of this analysis was to explore degradation rates, the data shows a few other interesting trends. First – the number of plants and total capacity deployed has increased significantly over time. In 2007, the first year of analysis, there were only 2 facilities totaling 22MW. In the peak year, 2016, the study shows 148 sites totaling nearly 10GW of capacity.

Also increasing – the DC (solar panels installed) to AC (inverters installed) ratio went from 1.18 modules watt per inverter rating, to a peak of 1.37:1 in 2018.

This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact:

Popular content

Transfer switch for home solar power integration
09 July 2024 Nature’s Generator now offers a 50-amp, 12-circuit switch to manually power up selected circuits from backup system.