UiO researchers have a goal: More solar and wind energy in the electricity grid
We waste enormous amounts of electricity from the wind and sun. Intelligent machines and large batteries will put an end to that Waste.
SOCIAL PROBLEM: We generate more and more electricity from the sun and wind. Unfortunately much of the most environment-friendly energy is lost because it is so difficult to regulate. Researchers have now started developing smart technology to overcome the problem. Photo: NTB Scanpix
Get into a car, drive 80 kilometres an hour day and night without stopping…for five months. The distance you will then have covered is equivalent to the length of the electricity grid in Norway. It is a formidable supply line – but, if used without storage, it has limited possibilities, with the generation of electric current continuously adapted to the amount we use: the amount of electricity that is fed into the grid must be exactly the same as what we take out.
“Grey, wet, no wind. That’s the weather today. But tomorrow? Maybe sun from a clear sky and a fresh breeze,” speculates structure physicist Sabrina Sartori at the University of Oslo, where we meet her under the glass roof of Forskningsparken AS.
Sun when we need it least
The sun and wind account for a steadily increasing amount of the electricity that we consume.
Unfortunately, this energy generation is difficult to regulate. It is slavishly dependent on the weather; when the wind slackens and the sunrays disappear behind the clouds, it is of no help to have installed capacity in the electricity grid. The output from both the solar cell panels and the wind turbines will drop dramatically regardless.
Fortunately in Norway we have large hydropower plants with reservoirs where the power generation can be rapidly adjusted up and down, whereas other countries have to rely to a greater extent on electricity from polluting coal and gas power plants when the sun and wind fail them.
Yet it is not just that the supply of resources varies; our energy needs vary as well – from hour to hour, from day to day, from summer to winter. The electricity from the solar cell panels is most plentiful when we need it least, in the middle of summer.
“A problem for society”
“Energy storage!” says Sabrina Sartori emphatically. “Energy storage will be extremely important in the coming years. If we are going to have any chance of solving the world’s climate problems, we will have to find ways to store solar and wind energy so that we can use that energy when we need it and not at the moment when it is generated. We can’t do that effectively at present. I think that is a major problem for our society…which we have set out to solve.”
Sabrina Sartori is an Associate Professor at the Department of Technology Systems and works with new systems for energy storage. She studies how the materials are structured atom by atom. Her research includes hydrogen storage in cars, and hydrogen systems and batteries intended for the storage of renewable energy.
“The left-over energy after a day with a lot of sun and wind ought to be stored in energy systems and used when there is a need for it. For instance, we need batteries with the enormous capacity that is needed if we are going to be able to efficiently transfer energy from solar and wind power over to the electricity grid. These batteries are currently under development.”
New research project
Sartori has just become the head of a new research project called INTEGRARE, which is Italian for “integrate”.
“The goal of this project is precisely to develop a much better integration of renewable energy sources into the Norwegian electricity grid.”
To be able to forecast the weather precisely and to make the grid capable of adapting to changes from hour to hour may turn out to be a key to greater use of electricity from the sun and wind.
“It’s difficult to plan the generation of energy from solar cells and wind turbines. We simply do not have the knowledge and expertise to do so,” she admits.
Precisely this important knowledge is what the researchers in the new project hope to acquire for themselves…and the society.
Must be able to forecast
“Our goal is to bring more energy from the sun and wind into the electricity grid while managing to maintain the balance between generation and consumption,” says Professor Frank Eliassen at the Department of Informatics, which is also participating in the research project.
To do that, it is necessary to be able to predict the magnitude of the generation of energy as accurately as possible in a given period – the next 24 hours, the next hour, the next fifteen minutes.
“Knowledge about the short-term variations in wind and solar intensity is important so that the supply to the electricity grid from these sources can be adapted to the constant variations,” adds Sartori.
In order to do that, of course, we need as accurate weather forecasts as possible: how strong the wind will blow, how intensely the sun will shine. However, it is not easy to determine the magnitude of the energy production from the state of the upcoming weather. We cannot do that yet because we do not know enough about the amount of energy generated by different types of weather. It is difficult to predict the future, so the researchers fall back on the past.
“However, not all methods are equally suitable for predicting future local energy generation. In this project, we will find methods that employ historical data about actual electricity generation during various weather conditions so as to get the best possible prediction. If we are able to make accurate predictions on the basis of historical data – where we have the result and can check whether we get it right – there is also good reason to believe that we can predict future energy generation,” says Eliassen, who has specialized in precisely the field of energy informatics.
The researchers in the project will get computers to make decisions that economize on important energy resources. More specifically, they intend to develop methods to enable computers to learn particular tasks on the basis of both historical data and experimental observations.
In order to do this, the researchers use what they call machine learning. They try to develop a mathematical, statistical scheme that can estimate the electricity generation for the next 24 hours, the next hour or the next quarter-hour when the weather forecast is issued for the relevant area. Of course, this requires that the weather report be correct.
Frank Eliassen thinks the project can result in better systems for prediction, which energy companies can employ. “I think these improvements can be the basis for a business concept,” he asserts.
An important objective for the project is to make the electricity grid flexible enough to receive the energy from solar cell panels and wind turbines.
”When we know how the weather will be tomorrow, we can adapt the use of the electricity grid to determine when we ought to use the energy stored in batteries and when we ought to use directly generated energy. This can be an effective way to prevent us from wasting enormous amounts of renewable solar and wind energy as we unfortunately do at present,” notes Sartori.
INTEGRARE is a so-called seed project. The researchers have high hopes that it will develop into a major European project. “For the problems we are trying to solve are found in all countries that are trying to develop the most environment-friendly power supply in existence – solar and wind energy,” says Sabrina Sartori.