Image Copyright:

The Fraunhofer Institute (IEE) in Kassel has developed a new forecasting system which is based on artificial intelligence.


In the process of switching to renewable energy, Germany requires ever more sophisticated forecasting methods for grid network operations. The Fraunhofer Institute for Energy Economics and Energy System Technology (IEE) in Kassel has therefore developed a new forecasting system which is based on artificial intelligence. GridFox can determine feed-in and consumption by individual plant or generation source and by consumer group.

Operators of transmission and distribution networks can use GridFox to regulate power needs and surpluses and make forward calculations that comply with the new redispatch guidelines (“Redispatch 2.0”) that will come into effect from 2021. The amendment to the Network Expansion Acceleration Act (NABEG) states that all operators will have to use all generation plants with an output of more than 100 kilowatts – including renewable energy plants – for redispatch measures.

"As a result, distribution network operators in particular will need significantly more powerful forecasting instruments," explains Dr Axel Braun, head of the Energy Meteorological Information Systems business unit at Fraunhofer IEE in a press release. In particular, the operators need to be able to locate and reliably calculate the occurrence of “vertical power flows” (the sum of feed-in and consumption) at individual transformers hours or days in advance. "Above all, however, they need reliable information about how much electricity will be generated by which system and when, in order to be able to regulate the feed-in in the redispatch in a forward-looking manner,” says Braun.

To predict energy flow per source, GridFox uses numerical weather forecasts and master data from the network operators. It also takes price forecasts and calendar data into account. Critically, the intelligent system enables the operators to generate the forecasting calculations themselves.