Solar panels have a love-hate relationship with nature and should be installed in exposed areas in the sun, but on cloudy days production is clearly reduced. Less obvious, more extreme weather from snowstorms to hurricanes can completely damage or destroy solar hardware. New research Performed by Sandia National Laboratories and published in Applied Energy, it shows how meteorological events can reduce US solar energy production.
To study this relationship, researchers have introduced machine learning algorithms into large datasets from private solar power plants. “This was a large collaboration,” Thushara Gunda, one of the authors of the treatise and a researcher at Sandia, told Ars. In the future, Gunda hopes to extend this research to investigate other extreme weather events and renewable energies such as wind, geothermal and ocean energy. She said her team was in the early stages of this work.
The team hopes that this study will help make decisions about future PV operations. This is especially true as climate change can increase the frequency of extreme weather events and cause problems affecting solar power. “We recognize that the transition to renewable energy has increased our reliance on local environmental conditions,” Gunda said.
First, Gunda and her co-author, Sandia researcher Nicole Jackson, have collected more than 800 maintenance tickets from solar farms in 24 states. Then they went back and forth with the solar company trying to understand the dataset. For example, different companies sometimes used different terms for the same thing.
Some companies have classified snow events and hurricanes as storms in their maintenance tickets, so the team needs to do a lot of analysis to determine what each company means when using the term “storm.” had. “No doubt, in terms of industry and day-to-day operations, the storm can be a sunny day,” Jackson told Ars.
“Just because someone wants to share data doesn’t mean they can analyze it automatically. There are subtle differences in how data is collected,” Gunda said.
Researchers have also obtained more than two years of power generation data from more than 100 solar power plants in 16 states and historical meteorological data for those areas. From there, the author ran a machine learning algorithm on the dataset to investigate the relationship between energy production and bad weather. This algorithm allowed the team to identify points where weather-related power drops match maintenance tickets and many other variables.
Shed light on the situation
The team found that snow events caused the greatest decline in performance (54.5%), followed by hurricanes (12.6%), and generally storms (1.1%). Somewhat surprisingly, hurricanes were mentioned in almost 15% of maintenance records. Other factors that lead to poor performance include plant size, age, and location. “We found that older farms were more likely to be affected by performance issues,” Gunda said.
However, there are subtle differences in this task. First, just because older sites are more affected doesn’t necessarily mean they’re less productive. Rather, they are only exposed to more weather than younger sites (even older farms are relatively young, 3-5 years old). In addition, the sites where researchers collected data were biased towards North Carolina and California. These states were prone to severe weather events that may not occur in other parts of the United States.
The team was also surprised that neither hail nor wildfires were included in the data. This does not mean that these events are not occurring. There are many fires on the west coast. Rather, these events weren’t specifically included in the maintenance ticket, as they are created only when the company has something to do. Given that the hailstorm is covered by insurance, these events may appear in the insurance database.
“But from conversations with the industry and attendance at conferences, we know that these particular events are certainly interesting,” Gunda said.
Amount of money
Insurance data certainly tells the story when it comes to how much damage a solar farm can face during a hailstorm.Report from National Renewable Energy LaboratoryPublished last year, it uses data collected from insurance service company Verisk to investigate the amount of damage a weather event can cause to operate solar power. (Insurance data also includes figures for vandalism and theft).
Data collected between 2014 and 2019 suggests that hail caused the most claims on solar hardware, with 7,979 cases and an average cost of $ 2,555. “Hail is a big problem for solar panels,” NREL senior researcher Andy Walker told Ars.
Fires were less common (1,282 cases), but the average bill was much higher, at $ 17,309. There were 79 freezes, including ice and snow, averaging $ 5,288. However, these averages include the cost of both commercial and residential PV projects. For example, in the case of a home freeze, the average bill was $ 4,195, while in the case of commercial activity it was $ 32,964.
Unpublished NREL studies also suggest ways solar panels can withstand extreme weather, Walker said. Methods include watertight enclosures, modules mounted on three rails (rather than two), thicker glass, windbreak fences, marine grade steel, and through bolts (rather than clamps). “The clamp turned out to be a smoking gun in many module releases. [photovoltaic] The module blows the rack away, “he said.
The upgrade costs a few cents per watt, Walker said. Some of these methods are useful for the various weather events that solar panels see and threaten to survive, such as being crushed by excess snow, blown off racks, or struck by hail. Increases the size of. “Solar panels are one of the most exposed in the build environment,” he added.
https://arstechnica.com/?p=1792683 Relationship between solar power generation and weather storms