How AI can help combat power grid congestion

AMSTERDAM: Vehicles in this Dutch city is putting unexpected and considerable pressure on the power grid, particularly during peak hours.

This is why a smart charging trial has been launched to intelligently regulate vehicle recharging without negatively affecting the local grid.

This grid congestion limits the city's ability to build new homes. In fact, it would be necessary to systematically review the electrical infrastructure so that the local network could adapt to ever-increasing needs.

To remedy this problem, the city of Amsterdam has launched a pilot project in collaboration with Deftpower, an AI-based software-as-a-service platform for electric vehicle charging.

Its platform already enables users to optimise their charging sessions according to hourly electricity rates and grid availability.

Here, the aim is to monitor the battery charge state and the scheduled departure times of all connected vehicles in real time.

This makes it possible to optimise each car's charging schedule according to individual needs and grid capacity.

Thanks to this approach, 60-70% of charging consumption has been shifted to off-peak hours, reducing pressure on the local power grid. For Deftpower's CEO, Jacob van Zonneveld, “one public charging station equals four houses” in terms of demand on the grid, according to Electrive.

The idea is that by reducing energy demand during critical periods, the city will be able to more easily approve future construction projects, without necessarily having to invest in grid reinforcement. In this way, smart charging could contribute to partially alleviating the housing crisis.

This pilot project is being carried out in collaboration with Equans (charging infrastructure operator) and the ANWB (mobility services provider).

If it proves conclusive, it could be extended nationwide from as soon as this summer, and could inspire other major European cities facing similar challenges in terms of energy requirements and housing shortages.