The past few years have shown that many areas are insufficiently prepared for severe weather events and other disasters. This stems from a complex skein of challenges, including underfunded or outdated systems, lack of clear authority to act, human errors and many other issues. Speed of response is clearly important. If a developing emergency can be acted on quickly, faults can be mended, outages rerouted, and probably lives saved.
With this in mind, Greece’s Center for Research and Technology, called CERTH, has developed an algorithm which will maximize the autonomy of the smart grid under emergency scenarios. This innovative system was created under the EU-funded microgrid project TIGON, which assigns priority to renewable energy sources (RES) and makes use of AI neural networks to provide forecasts of production and demand for renewable power available.
The researchers believe that smart grids – which can handle demand under normal circumstances – will be challenged when an emergency disrupts the system. The operating software of a future smart grid would need to be able to respond effectively to various kinds of emergencies.
CERTH's researchers created a software algorithm model, where an optimizer responds to emergencies, and excludes faulty components from the network, then uses the remaining smart grid measurements as input, deploying solar power, wind production and battery storage to make up for the loss of capacity.
The researchers modeled a variety of different emergency scenarios to see what impact loss of say, wind generation power at different times would do to the rest of the network. The conclusion was that a mixture of different systems showed the best resilience under disaster conditions. This might seem obvious but it is important to ensure that disaster response mechanisms are built into advanced smart grid systems so they are not overwhelmed by Black Swan-type unexpected events or failures.