The danger of overloading the electrical network could soon hit Germany. In Norway this was already the case. Solar cells on the roofs and heat pumps load the electrical network, but that's nothing compared to the pressure that electric cars will put on the net when they seriously make their entry. "Intelligent energy management systems for electric vehicles can safe many investments in the utility grid infrastructure and make a more efficient use of local energy resources possible," says Robert Brehm, scientist and head of the German-Danish Interreg carpeDIEM project. In the end, the consumer saves costs, because usually the investments for the expansion of the supply infrastructure are redistributed among the customers. In the carpeDIEM project, DIEM stands for Distributed Intelligent Energy Management.
- Did you talk to your car today?
A brief rewriting of the slogan, "Have you talked to your child today?", might contain the solution to the challenge the power grid faces today. The challenge is not only that private citizens have become small electricity producers with roof-mounted solar cells and heat pumps, but also the goal of governments to phase out fuel and diesel cars by 2030. Electric cars are huge power guzzlers.
That by itself is not problematic, yet. It turns into a problem when a large number of electric vehicles within a supply network are charged simultaneously. An example is residential areas in which workers, who drive to work by e-vehicle return home in the evening at peak times and then simultaneously charge their e-vehicles for the next use.This then results in a rush hour in the supply network. Unfortunately, today's electrical infrastructure is not designed for such high peak loads. For comparison, the largest household consumers, such as washer and dryer have a maximum connected load of 2 to 2.5kW, while an electric vehicle charges today with power from 11kW to 22kW. The result of an inadequate supply infrastructure with a simultaneous increase in the number of e-vehicles could already be seen in Norway and California with blown cables and power outages.
There are two solutions. Either we replace the power grid or we do something smart. And then we're back to the question of whether you've talked to your car today.
Prof. Brehm from the Mads Clausen Institute in Sønderborg together with the project partners of the University of Applied Sciences Lübeck, the Europa Universität Flensburg and CBB Software GmbH, measured and analyzed the energy consumption data of a municipality in Nordfriesland. Within the project carpeDIEM, they developed an intelligent load management system for the cooperative charging of electric vehicles. Core of the system are charging stations and charging boxes for electric vehicles, which communicate with each other to integrate as much as possible local energy resources and to prevent network load peaks by coordinated charging.
The device can thus speak with your electric car. And it works.
The solution saves major investments
In spring 2018 the project team around Prof. Brehm investigated the usage profiles of the car fleet of a network partner, 'Kreisverwaltung Ostholstein' in the state of Schleswig-Holstein. The administration seeks to be prepared for the future and plans to replace their company cars with electric vehicles in the long term. The future charge management however poses a challenge.
To be able to analyze the case technically and in respect to peak loads the researcher in Sønderborg received logbooks about the use of the vehicle fleet. The statistical analysis revealed that 90 % of the cars returned between 16-18 o’clock with less than 100 km driven. The usage pattern of the diesel-powered fleet has now been projected onto the use of electric vehicles and the resulting network load for the electric vehicles has been simulated over time. Figure 1 shows the resulting network load for one day. Clearly recognizable is a peak load of more than 200 kW, which would strain the transformer station and the overlaying network around 5 pm. "The figure demonstrates how the maximum capacity of the supplying substation is clearly exceeded," says Prof. Brehm.
Fig. 1 Simulation of the load behavior of a single day, when all electric vehicles are charged directly after arrival.
One possible solution is to set static limitations on charging stations so cars could maximum load at 10 kWh. This will mean an increased charging time, but results in a spread of the load over time with some remaining load peaks. A more advantageous solution to the challenge would be dynamic charging. Here charging of the cars is dynamically distributed over different vehicles. The starting point for this is the dwell time of the respective vehicle at the charging station - this information is always given in the example because the vehicles are only needed for work next day. Now, the total amount of energy needed to fully charge all vehicles by the next morning will be split over the available time period and coordinated for all vehicles. The simulation clearly shows that this approach leads to very smooth profiles at a significantly lower level (see Fig. 2).
Fig. 2 Dynamic charging of all cars on a single day, mind the enlarged scale of more than a factor 10 compared to Fig. 1.
The team around Prof. Brehm is currently working on a charge station that is able to learn using artificial intelligence.
This example shows what you can achieve by intelligently controlling the charging of cars and at the same time save money.
The carpeDIEM-project is financed by Interreg Deutschland-Danmark with means from the European Regional Development Fund.