Streamlining the energy transition with mathematics and informatics

The large-scale transition from fossil to renewable energy sources has major implications for the electricity grid. On the supply side, greater fluctuations arise, for example in solar and wind energy. Moreover, supply is becoming less dependent on a few large energy suppliers and more decentralised. Individuals are increasingly becoming energy producers themselves by generating power through their own solar panels and feeding it back to the grid. Moreover, sources of alternative energy, such as offshore wind farms, are often far from industrial centres. Therefore, new high-voltage grids are needed, even twice as much between now and 2050.

On the demand side, the need for electricity will increase significantly, for example as transport and home heating become more and more electric, and as gas is replaced by electricity in industry.

Probability of blackouts

What should the future electricity network look like to cope with all these challenges? To answer that question, CWI is conducting different types of fundamental research in the area of mathematics and computer science, in order to contribute to challenges in the electricity network.

For instance, CWI mathematician and professor at TU Eindhoven Bert Zwart is investigating the probability of power outages in the electricity network. Such 'blackouts' range from local to national and even international. Zwart discovered in 2020 that the probability of power outages is determined more by the distribution in the size of cities than by the details of power grids. With a better understanding of the causes of blackouts, engineers can design more robust power grids.

It is obvious to think that the energy transition will have a negative impact on the number of blackouts. After all, there will be greater fluctuations in supply and demand. Yet Zwart sees it differently. He thinks that with an adapted design of the power grid, the number of blackouts could even decrease.

"Neighbourhoods in Adelaide, Australia, were left without power for half an hour in turn for weeks in 2016," Zwart explains. "People had switched from coal-fired power plants to renewable energy, but only started thinking about the impact on the grid when the coal plants were already closed. It is not obvious to blame renewable energy for the blackouts then. It was politics that made changes too quickly."

Zwart compares the transition from our current centralised electricity grid to the future decentralised electricity grid with the transition from the analogue telephone network to the digital internet. Zwart: "That transition was made possible by the TCP/IP protocol, which sends digital information in packets in a smart way."

The TCP/IP protocol cuts up a data stream into packets and sends them as in a relay from one intermediate station to another until they finally arrive at the recipient, sometimes even via different routes.

"The holy grail for our future energy network," says Zwart, "is to develop something like TCP/IP for energy networks: an algorithm that sends packets of energy around smartly. The extent to which we can develop a similar smart algorithm for energy will determine the extent to which we can digitise the energy network."

Trade off between efficiency and robustness

Energy transformation is forcing us to become more self-sufficient and come up with more local solutions. Zwart draws another comparison with the internet: "If the internet is too busy, data packets are stored in routers. You can do something similar with the electricity network by, for example, putting a large battery in each neighborhood of a thousand inhabitants for when the network cannot supply electricity temporarily."

The future electricity network must be able to respond flexibly to changes, says Zwart. "If we manage to do that, then the chances of blackouts we have in our current grid may actually decrease. But it's always about finding a balance between efficiency and robustness. More efficiency means less robustness, and vice versa."

Intelligent agents trade energy

An energy system with consumers who are simultaneously producers also changes the ways in which energy can be traded. While consumers now buy electricity from a few large providers, they will in principle be able to trade electricity among themselves in the future. But that has to happen automatically, otherwise it will not be practical.

In that context, CWI researcher and professor of Intelligent Energy Systems at TU Delft Han La Poutré, together with several PhD students, is studying mechanisms to achieve flexible energy contracts between different parties. "Flexibility in contracts," says La Poutré, "can contribute to a solution for greater uncertainty in the supply of renewable energy on the one hand, and to a solution for congestion in the grid on the other."

Flexibility in contracts can take the form of micro-contracts between two parties. These so-called peer-to-peer contracts deal with price, quantity and security of supply. La Poutré: "For such micro-contracts, we develop automated negotiation strategies based on game theory. We can then prove certain properties of that mathematically, so we have certainty that it works well."

The developed negotiation strategies can be translated into algorithms, and these in turn form the basis for smart, autonomously acting 'agents', say pieces of software. Those agents can automatically negotiate between two parties, for example between a consumer and a wind turbine, or between two neighbours. The people themselves do not have to do anything because it happens automatically.

Apart from mechanisms for flexible contracts, La Poutré is also investigating fairness in congestion management. La Poutré: "Then the question is how much you can cut off private individuals and companies from electricity when there is too little to supply and what is a fair distribution between them? This is very relevant, as we already have about six thousand companies in the Netherlands that cannot be connected to the power grid. For different ethical standards, we are trying to develop algorithmic solutions and prove that they do what they promise to do."

La Poutré has chaired or co-chaired defining NWO programme calls on energy four times, especially on the integration of energy systems (such as combining electricity and hydrogen or heat). As such, he has helped set the direction of the content of all NWO programmes on energy.

New power switches

A very different kind of research that contributes to streamlining the energy transition is done by CWI physicist and applied mathematician Ute Ebert, who is also a professor at TU Eindhoven. Looking for a new type of power switch, she and her colleagues are studying how electrical discharges in a gas arise and can be stopped.

"Every high-voltage grid uses circuit breakers, which control the currents in the grid," Ebert explains. "In the current grid, circuit breakers contain the gas SF6, which is almost irreplaceable, but unfortunately is also the world's worst greenhouse gas, more than twenty thousand times worse than CO2. During maintenance, some of this gas constantly escapes. The EU has banned the use of SF6, except in these switches. The EU does require the development of a more environmentally friendly alternative within ten years."

When a circuit breaker needs to cut power, two metal electrodes are pulled apart. However, the electricity then still tries to flow through the gas between the electrodes. The gas should prevent this, or, if it does happen, it should extinguish the electrical discharge as quickly as possible. The big question is which gas offers the best alternative to SF6.

Ebert and her colleagues have built up a unique expertise in simulating such gas discharges in great detail on the computer. "For gas discharges in air, we have recently shown that our simulations agree with experiments. For this, we work together with electrical engineers at TU Eindhoven who can generate extreme voltage pulses, and with physicists who can measure exactly what happens in the air."

She also wants to develop the computer simulations that already work so well for air as a medium for possible new gases for the power switches, but that is not an easy task. Ebert: "The gas discharge depends on the geometry of the switch, the voltage pulse and the composition of the gas. We already know that simulations in gases other than air present us with new challenges. But we are happy to take them on to understand how such a discharge occurs in the new gas and how to get rid of it as soon as possible."

The work of Bert Zwart, Han La Poutré and Ute Ebert are three examples of fundamental mathematics and computer science research through which CWI is contributing to the energy transition. The text below in the yellow box refers to some other CWI projects in the field of energy, sustainability and energy transformation.

Read more:

• City sizes may affect blackout probabilities: https://physics.aps.org/articles/v13/122
• Emergence of scale-free blackout sizes in power grids: https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.125.058301
• (Dutch only) Hoe onze obsessie met efficiëntie ons kwetsbaar maakt voor onwaarschijnlijke gevaren: https://decorrespondent.nl/11191/hoe-onze-obsessie-met-efficientie-ons-kwetsbaar-maakt-voor-onwaarschijnlijke-gevaren/988256674515-20cde0d8

Text: Bennie Mols