The more electricity comes from renewable sources, the more complex the electricity grid becomes. Energy computer scientists at the University of Oldenburg are researching the transformation to a flexible and intelligent system as part of a DFG Priority Programme.
Bottlenecks, high costs and an unstable supply - many consumers associate these problems with the expansion of renewable energies. In fact, the energy transition poses major challenges for electricity grids. "In future, there will be many small, decentralised power generators. In addition, the energy conversion of renewable sources such as wind and solar power will fluctuate," reports energy computer scientist Prof Dr Sebastian Lehnhoff.
Together with colleagues from other German universities, Lehnhoff is investigating various aspects of the transformation of the electricity grid into a "smart grid" that intelligently connects consumers and producers. The computer scientists at the University of Oldenburg are involved in three projects within a priority programme set up by the German Research Foundation (DFG). They are each being funded with almost 300,000 euros and have recently been launched.
Decentralised control
One of the three projects involving Oldenburg is looking at how the numerous renewable electricity sources - such as photovoltaic systems, wind turbines or combined heat and power plants - can best be coordinated. Lehnhoff and his colleagues from the University of Hanover are using what is known as a multi-agent system for control purposes. This is software that works in a similar way to an ant colony: local control units act independently of each other. They ensure, for example, that the voltage in one area of the network remains within the permitted range without a central centre monitoring the entire system. "This automatically creates redundancies that make the system more robust and less prone to errors," says Lehnhoff.
In the second sub-project, the Oldenburg researchers are working with colleagues from TU Dortmund University to investigate the risks posed by the power grid's dependence on information and communication technology. They want to develop a model that detects and evaluates security-critical points. In this way, the team wants to find out whether failures can spread in a cascade, for example from the electrical grid via the telecommunications network to the heating and gas network. "The final step is to create a risk analysis to identify the probability of critical events occurring," reports Lehnhoff.
Powering up after the blackout
The third sub-project deals with the so-called black start - i.e. the case where the grid has to be restarted after a large-scale power outage. Here, the Oldenburg energy computer scientists are working together with the University of Passau. "A black start requires careful coordination between the IT systems and the electrical energy system," explains Lehnhoff. A classic black start in a large power plant begins with a battery that starts the control system for a small diesel generator, which in turn starts a gas turbine. Hydropower plants, whose turbines can be set in motion without electricity, are also frequently used.
In future, however, the task will become more complex: in order to rebuild a stable electricity grid, many decentralised producers and consumers will have to be coordinated. The problem is that modern information and communication technology is needed to get such "smart grids" up and running after an outage - but this technology itself places high demands on grid stability. "In the event of a large-scale system failure, IT and the electrical grid must be rebuilt and restarted in parallel, with the two systems interacting dynamically with each other," reports Lehnhoff. The computer scientist wants to solve this "multi-criteria optimisation problem" within the DFG project.