How AI could help Switzerland switch to renewable energies
Switzerland wants to rely exclusively on renewable energies by 2050. The construction of more solar panels and wind turbines plays a decisive role here – as does artificial intelligence.
This content was published on December 2, 2021 – 9:00 am
Katherina alarm clock
Charge the smartphone, do the laundry or stream a series on Netflix – we usually don’t have to worry about where our electricity comes from. Or whether it’s enough to power all of our devices.
Utility companies and researchers have also not given much thought to supply in the past. However, since Switzerland is increasing its share of renewable energies, the question of the need-based supply of electricity is becoming more and more important.
First, the sun doesn’t always shine and the wind doesn’t blow all the time. Second, the demand for electricity is changing. Since we’re leaving fossil fuels behind, most new cars run on batteries. Heat pumps will replace oil and gas heating in buildings.
Third, the energy mix is evolving. Around 76% of the electricity from Swiss sockets comes from renewable sources, mainly from hydropower plants, while the share of nuclear energy is 20%. In Switzerland there are only a handful of large companies that operate hydropower or nuclear power plants. These companies know exactly how much electricity they are generating at what time.
But nuclear power is gradually being phased out and replaced by renewable energies. Nowadays, any household with solar panels on the roof can potentially generate electricity. Small and large wind farms across the country feed into the power grid. This changes the way the network works.
“We are moving from a central to a decentralized system, which makes the operation of the power grid more complex,” says energy expert Matthias Eifert. He founded Energy Data Hackdays, an annual event where hackers, data analysts and engineers spend two days working on solutions to accelerate the energy transition. The greatest challenge, according to Eifert, will be to ensure that the new generation locations are efficiently connected to the grid.
One solution that he advocates is the use of artificial intelligence. “Machine learning and artificial intelligence can help to optimize and stabilize the interaction between energy supply, use and storage,” he says.
Eifert is not alone in making this assumption. The World Economic Forum sees “enormous potential” in artificial intelligence (AI) to “accelerate the global energy transition”. In a white paper published in September, the organization called on governments and companies to invest in AI.
So how exactly can AI support the transition to renewable energies?
Energy stored in electric car batteries could power households
The first step is to understand people’s behavior and to understand where potential discrepancies between supply and demand could arise, says Ben Bowler from the Lucerne University of Applied Sciences and Arts. He is a senior researcher in the field of digital energy and has been dealing with energy storage and grid infrastructure for years.
It may seem surprising, but utility companies usually know very little about people’s energy use. You know how much electricity each household uses over a certain period of time. But they do not know on which days and at which times the electricity consumption is particularly high or low.
A project by Ben Bowler focuses on analyzing energy data and understanding human behavior – with the help of an algorithm.
It collects the data from so-called smart meters, which are being rolled out across Switzerland and replacing the old basic electricity meters. According to the Federal Office of Energy, 80 percent of households should be equipped with a smart meter by 2027. These meters monitor household usage in real time and report back to the electricity provider every 15 minutes. This way, companies know when it is peak time, such as when people turn on the dishwasher after dinner.
“We try to predict people’s behavior. We look at electricity consumption so far and try to predict what people will do tomorrow. Based on the data, we can understand whether there could be problems on the network, ”says Bowler.
Understanding human behavior better is part of the equation. The other part is finding ways to make sure there is enough power available during these peak times.
This is where another Ben Bowler project comes in. He is investigating how batteries from electric cars could be used as short-term storage for the power grid.
As electric car sales rise, so will pressure on the power grid. In order to avoid supply bottlenecks, the charging of cars is to become a one-way street. When the electricity demand is low, the cars are charged at intelligent charging stations. And when demand is high, the energy stored in the car batteries can be discharged and fed back into the grid.
The pilot project of the Lucerne University of Applied Sciences is a cooperation with the energy company Tiko and the start-up Sun2Wheel and is being tested with 50 vehicles from the car sharing company Mobility.
“It all depends on getting better data on how people use the cars, when the sun is shining, when the cars are charging. Then machine learning and artificial intelligence analyze the data, ”says Bowler. Researchers are working on similar pilot projects with other car sharing companies in Germany and Denmark.
There is still a long way to go before vehicle-to-grid technology goes from testing to implementation and ultimately to viable business. There are not yet many electric cars that allow bidirectional charging. Nissan, Volkswagen and Fiat are among the few brands. Mobility currently only has 150 electric cars in Switzerland, but plans to electrify its entire fleet – more than 3,000 cars – by 2030. What is then also needed are smart charging stations, such as those developed across the country by the Swiss start-up Sun2Wheel.
Anonymized personal data
The challenge that Bowler and his colleagues encounter during the process is how to use the data – what time people use electricity, when they do laundry or charge their electric cars – without violating privacy laws.
A start-up from the USA believes it can help. VIA Science, which recently opened an office in Zug, has developed a program that is being tested in the smart meter project at the Lucerne University of Applied Sciences. Instead of having to extract the data, the researchers can evaluate it directly on the meters. Private personal data doesn’t need to be sent anywhere.
There is also another solution to privacy issues: encouraging people to voluntarily share their data. Dozens of scientists, hackers, students and representatives from energy providers met at the Energy Data Hackdays in Brugg in September to consider how data can contribute to a stable, climate-neutral energy system.
You launched the “Read your own smart meter” project. This would allow homeowners to visualize the data on their smart meters and give them insights into how much electricity they are using and where they could potentially save electricity – and money. They can then share the data anonymously with the utility companies.
Switzerland is a leader in artificial intelligence. It is also one of the countries with the strictest data protection laws. If you can solve privacy issues, “people can help the grid and support decarbonization,” Bowler says.