SALM
Self-adaptive charging management - AI for electric charging infrastructure
- Simulation is essential – The more complex the charging system, the more important realistic simulation models are.
- Key figures for comparisons – charging infrastructure must be analyzed using measurable KPIs such as capacity utilization, charging quality and energy consumption.
- Transparent visualization – Operators need comprehensible representations at the right level of detail in order to be able to make decisions.
- Electromobility needs balance – the ramp-up will only succeed if infrastructure expansion and user needs are in balance.
In the SALM project, real-life situations are simulated in order to effectively and efficiently investigate the optimization of charging strategies. Optimizing the charging infrastructure with artificial intelligence is an important contribution to climate-friendly mobility.
For years, we have been optimizing the flow of electricity for electric vehicles with Gridware – cloud-based and intelligently controlled. In the SALM project, together with the University of Kassel and the House of Energy, we have been researching how artificial intelligence (AI) can make charging more efficient.
Challenge & solution
As the number of e-vehicles increases, so does the load on the electricity grids. While some drivers need fast charging with high power, others have more time but higher energy requirements. SALM optimizes the charging process through the use of artificial intelligence (AI), which intelligently controls the entire energy system consisting of the grid, charging stations and vehicles.
A central element is the “digital twin”, a simulation of the charging infrastructure. This model enables ongoing optimization of charging strategies and helps to make the use of the networks more efficient.
How we continue to develop gridware
We integrate the findings from SALM into Gridware to provide the best possible support for charge point operators (CPOs):
- Key figures such as utilization, charging quality & energy consumption are available for all hierarchy levels
- Algorithms are continuously optimized to control charging processes more efficiently
- Simulations of large AC & DC charging clusters help with the planning & expansion of the charging infrastructure
- AI should help to plan the gradual expansion of the networks with foresight
Funded by:
