Solar panels learn to “dance with the wind” to stay strong in the face of storms

Researchers at the Center for Materials Forming at PSL University in France have combined artificial intelligence (AI) and machine learning with computational fluid dynamics to help protect solar panels from extreme winds.

The technology is designed to help minimize downtime of renewable energy resources in the face of increasing extreme weather events.

In the fight against climate change, photovoltaic or solar panels constitute an important tool. Capable of transforming abundant sunlight into electricity, these panels can help produce large amounts of clean energy and reduce our dependence on fossil fuels.

Countries around the world are investing in solar power plants to meet their net zero emissions goals, making solar power the fastest growing sector in the energy sector. However, as the planet warms, we are seeing an increase in adverse weather events, and the reliability of this renewable energy resource is a major risk that we must prepare for.

Wind: friend and foe

Blowing winds are both good and bad for solar power plants. When blowing slowly, the winds help remove dust and dirt from the surface of the solar panels. This allows the panel to receive sunlight across its entire surface and maximize energy production.

Additionally, the blowing winds also serve as cooling agents for solar panels. During operation, photovoltaic panels tend to accumulate heat internally. This reduces their energy production efficiency. The winds help reduce heat buildup inside the cell and keep the panels operating efficiently.

To generate large capacity solar power plants, photovoltaic panels are installed over large areas of land. But it also allows winds to blow unobstructed, and when wind speeds increase, thin panels become very vulnerable to damage.

When disrupted by high winds, solar power plants can take weeks to recover from structural damage, which also disrupts energy availability.

AI to the rescue

To minimize damage to solar panels from high-speed winds, researchers worked on parameters such as ground clearance, tilt angles and row spacing. Even tracking mounts designed to maximize power generation by following the path of the sun have been repurposed to minimize damage in windy conditions.

A solar power plant can direct its panels to assume a safe storage position by remaining parallel to the ground when winds become strong. However, this is not enough when strong winds blow.

To solve this problem, a research team led by Elie Hachem, professor at PSL University and head of the Computer Science and Fluids research group, turned to machine learning and artificial intelligence to create smarter panels able to make independent decisions to minimize damage.

“By combining advanced fluid dynamics and artificial intelligence, we saw an opportunity to innovatively address the risks of wind damage and contribute to the resilience of renewable energy systems,” Hachem said in a statement from press.

The researchers used machine learning to simulate wind conditions and optimize the angles of the solar panels against strong winds. Using the available data, the algorithm designs creative solutions to reduce stress. However, instead of telling the panel what actions to take, the machine learning algorithm allows them to become the decision maker and has found that it outperforms current protection measures.

“It’s like teaching panels to dance with the wind, minimizing damage while protecting energy production during high winds,” Hachem added.

Interestingly, the solution goes against the conventions of engineering practices, but is highly scalable and could help build resilient systems for a greener future.

The research results were published in the journal Fluid Physics.