GridnAltion
AI enhanced power to heat to power energy storage in district heating grids.
Challenge
The stable electricity supply of the future
The energy transition is decentralizing power generation, replacing traditional centralized supply. However, the grid isn't designed for rising demand and fluctuating renewable energy inputs. Therefore, smart innovations to enable flexible load management to ensure resilience of the system are needed.
Team
Kilian Golinski, Henry Klein, Alper Kinaci, Bikram Dutta
About the prototype
To enable district heating operators to join flexibility markets to support load management and grid stability they must be able to deliver the full requested power within five minutes. Due to thermal inertia a power to heat to power system is not able to reach this limit. With our Deep Learning Tool, we want to predict the time of activation signal for such energy storage systems, so that grid operators are able to start their thermal engines prior to the signal of the grid operators. Therefore, we want to use open-source data such as weather, grid, historical and operating data to train our tool. This will allow district heating operators to use their infrastructure as power to heat to power energy storages to participate in load management of the German electricity grid..
Outputs
Further information about the progress, milestones, and roadblocks.
Impressions
