Column by Jacob Østergaard, Professor and Head of Division at DTU Wind and Energy Systems. Published on Ingeniøren on 17.01.25.
When we talk about the transition to a 100% renewable energy system, we often talk about it as something that is about to happen. It is common knowledge that things are moving in that direction, and Denmark is one of the countries leading the way.
When we talk about the challenges associated with this transition, a point of concern is that there may not be sufficient green power at the right times. Sunshine or wind is not always guaranteed, and Energinet, where I am a board member, recently announced that the average outage time will increase from 22-30 minutes per year in 2021-23 to 36 minutes per year in 2034.
Research into this problem is well underway, and a number of dedicated, consumer-oriented solutions have been developed to curb it. Among the best known are probably the many smart charging solutions for electric cars that ensure that the car is charged when the price is low. That is, when the sun is shining and the wind is blowing. In addition, work is being done to make a number of other consumption technologies flexible so consumption can follow the fluctuating electricity production.
Something we very rarely talk about, however, is how we can succeed in designing and managing the energy system to ensure continuous stability.
Unknown territory
Right now, we don't know enough. This is a problem, because if we do not know how to design the overall system to ensure stability, and get a common language for the control, we do not know how to design the individual system components. Or how to better safeguard the system against crashes when we are exposed to cyberattacks.
Specifically, it is about learning more about how a system based on power electronic converters can be controlled digitally. The most important task of converters is to quickly and efficiently convert direct current to alternating current and vice versa, so your car battery can be charged with alternating current from the electricity grid, for example.
An increasing share of the future electricity consumption will be connected via converters that will increasingly be used in the grid for, e.g., large direct current connections connecting countries.
Production units such as solar and wind farms are also based on converters. They replace the large, central power plants that we know today and which so far have stabilized the energy system. The process at the power plants has been that steam is poured into a turbine, which causes a shaft with a large alternator to rotate. The generator is connected directly to the power grid and has a built-in inertia that ensures that imbalances and disturbances do not propagate in the grid immediately, but are absorbed by the generators at the large power plants, which creates stability.
The same will not happen with the converters, which do not have the built-in inertia of the generators, but are digitally controlled. So it is the way electronics are digitally controlled that determines whether the stability of the system can be maintained.
It is a completely new way of designing an energy system. We are moving in new and unknown territory.
Energy infrastructure boosts development
Ensuring a future with a robust, renewable energy system that responds appropriately to errors and attacks will require a targeted prioritization of research and innovation of the energy system as such and of digital control.
Research into how to balance an energy system with a large proportion of wind and solar is underway. The experimental facility SYSLAB, which is part of PowerLabDK and linked to DTU Wind and Energy Systems at Risø, is currently being finalized. Until now, the facility has consisted of a mini-electricity system based on a renewable energy system, but currently district heating and hydrogen infrastructure are also being connected to form a fully interconnected, renewable energy system.
This enables us to develop solutions in a protected laboratory environment, where we integrate the different infrastructures and manage them optimally in relation to each other. This means that we can get more value out of the technologies - for example by storing the heat from the heat pumps - and coordinate everything so the energy is used optimally.
Technologies tested in this environment can then be moved to, e.g., Bornholm, which serves as a living laboratory for an energy system based on renewable energy. Test facilities such as SYSLAB and Bornholm allow us to develop new market-based solutions that ensure that charging of electric cars can be coordinated - not just according to how much wind and solar is available. But also according to the local network capacity which supplies the individual consumer.
If such solutions are not introduced, the high proportion of flexible consumption will clump together for a few hours, causing the local electricity grid to collapse.
So the possibilities are plentiful, but when it comes to the quick digital control of the converters, the existing test facilities fall short.
Digital twin as a solution
It is therefore crucial that we have a tool to examine the new phenomena associated with an energy system based on digitally controlled converters. To do this, we must be able to study the energy system in a much higher temporal resolution, which enables us to see how a system based on converters reacts at microsecond level.
This is necessary because the digitally controlled converters operate extremely fast. And if the control fails, it will happen at the microsecond level. We must therefore be able to understand what is happening and develop new solutions in the high temporal resolution. A bit like when you take a high-resolution photo: It is razor-sharp overall, but you can also zoom in on specific areas to take a closer look at the small details.
Specifically, this will be possible through a digital 1:1 model of the energy system. High-resolution digital twin. With the digital twin, we will be able to test solutions and strategies, and study how the overall system reacts in detail to different control designs. For example, we will be able to test how converter control should be designed and coordinated. It will also make it possible to test the system response to a cyberattack, and subsequently design it to be robust against that particular type of attack.
In other words, we are in the process of developing the solutions that make it possible to balance the energy in the system hour by hour and minute by minute. But we have yet to effectively solve the challenges associated with an electricity system based on converters and digital control, where things happen at microsecond level. We have a pretty good idea of what it takes, but lack the tools to ensure a viable design.
Talking about digital control may not be as tangible as talking about smart charging of electric cars and balancing production and consumption. But it is necessary to prioritize the area if we are to realize the green energy transition together and not compromise our security of supply.
