Rare events in sustainable energy grids simulated

The transition to renewable energy sources poses new challenges to the reliability of power grids. As renewable energy is very variable, rare combinations of factors might lead to instabilities and even black-outs. Wander Wadman of Centrum Wiskunde & Informatica (CWI) developed new simulation methods that can determine the probability of such rare events.

Power grid

The transition to renewable energy sources poses new challenges to the reliability of power grids. As renewable energy is very variable, rare combinations of factors might lead to instabilities and even black-outs. Wander Wadman of Centrum Wiskunde & Informatica (CWI) developed new simulation methods that can determine the probability of such rare events. Wadman defends his thesis on this topic on Thursday 18 June at the University of Amsterdam.

Renewable energy generators such as wind turbines and solar panels supply more and more power in modern electrical grids. Although the transition to a sustainable power supply is desirable, it poses several challenges to the stability of power grids. Renewable sources such as wind and sun are very variable, increasing the risk of current overloads and voltage deviations. Also, power generation is distributed over countless small turbines and solar panels, making steering mechanisms such as switching power plants on and off much less effective.

To provide a reliable power grid, grid operators need to be able to reliably predict rare events where the grid becomes unstable. In his thesis, Wadman investigates methods to quantify the chances for such rare events. Simulating these events is hard, as the probabilities are so low that conventional simulations would need to run millions of times in order to find a reliable answer, which is unfeasible. To circumvent this, Wadman introduced the mathematical equivalent of ‘save points’ in the simulations. By allowing the simulations to return to a save point when no rare event is found instead of starting all over, the workload of the simulations could be reduced by several orders of magnitude.

In the near future, the rare event simulations developed by Wadman will be implemented at DNV GL (formerly KEMA), with the aim of improving the stability of future power grids.

 

More information:
PhD Defense Wander Wadman