Charging Infrastructure for Electric Mobility

Battery-electric vehicles put a strain on the power grid, especially at the low-voltage level. Even when a few cars are being charged at the same time, high charging capacities can overload the cables or the transformer. On the other hand, electric vehicles offer comparatively high flexibility with respect to charging time, charging duration and power. Intelligent controls can make use of this flexibility by remote access by the network operator, the charge point operator or local optimization. Therefore, intelligent control of the charging processes can not only avoid power grid overload, but also facilitate a higher use of renewable energy. Furthermore, the power electronic systems and the batteries in electric vehicles can provide system services for the grid; for example, they can supply the balancing energy for the regulatory power market. The demands on networking and controllability of charging systems are increasing.

In the future, electric vehicles and their charging infrastructure will be controllable, securely integrated into the system via smart meter gateways, bidirectional and interoperable. We work with our customers to create charging solutions fit for the future.

Our R&D activities on "Charging Infrastructure for Electric Mobility" include:

• Grid-supported and Optimized, Self-powered Charging
• Communication
• Energy Recovery
• Controller Simulation and Load Profiles
• Grid simulation

Those who have data are clearly at an advantage. Our charge management system makes use of all data which is necessary to optimize the charging schedule: state-of-charge, charging target (SoC and time), PV forecast, load forecast, current performance at the connection point, charging targets of other vehicles, stationary buffer batteries. The software guarantees safe operation within the rated power and incorporates time flexibility options as well.

The main goal of our charging management is to integrate various control mechanisms for charge management into complex charging infrastructures. Our charging algorithms recognize the time shift potential of the singular vehicles during charging and discharging. Inefficiencies during vehicle charging are minimized by selecting the operating point, and the sum of the maximum capacities at a network connection point is increased by coordinating the charging schedules. As a result, cable reinforcement and investments in larger transformers can be avoided in most cases.

Featured in our charge management software is the prediction-based optimization of the PV self-consumption, by means of a stand-alone charge controller or a module in a home energy management system. Periods of surplus energy generation are predicted and used to charge the electric vehicle. Real-time monitoring of the energy output onsite at the house connection point minimizes the effect of any deviations in the forecast. These measures result in a significantly higher internal consumption rate.

Our charge management has been tested and further developed in many research projects in the field. We use standard interfaces, which we adapt to new requirements if necessary. Please contact us if you are interested in either the complete charge management system or an individual module thereof. 

In order to universally integrate electric vehicles and charging infrastructure into the distribution network, a connection with standardized communication protocols is required. We work with all protocols that are currently relevant and develop these further, adapting them for new control concepts and applications. We offer services ranging from the vehicle to the charging station and up to the charging management backend.

You can find further information at  www.openmuc.org

The power electronic systems and the batteries of the vehicles can benefit the power grid by providing system and grid services. Excess energy can be stored in the vehicles as needed and discharged when required. This flexibility option not only avoids expensive load peaks during times of short-term, high-energy demand but also increases the share of renewable energy use.
Our charge management algorithms take into account the economic efficiency of the energy recovery and monitor the complete charging process. Together with our customers, we develop and test bidirectional charging infrastructures in our Digital Grid Lab.

A detailed simulation environment is necessary in order to develop and evaluate load management algorithms. The Fraunhofer ISE load profile generator “SynPRO” is an established tool for generating time series data for household and commercial loads. This data can be used to simulate the interaction of non-controllable loads and optimized charging processes.

To successfully optimize charging processes, we provide forecasts of the remaining (non-controllable) loads. We calculate these using machine-learning algorithms based on the system measurement data. Different methods are used and then consolidated into a final prognosis by ensemble techniques. This keeps the uncertainties as small as possible.

The models and data are used to intensively test and evaluate charging strategies before they are implemented in the field. As a result, we can accurately determine the performance of a charge management algorithm and qualify its potential.  Also, we can provide load profiles of the system at the grid connection point. These can then be incorporated into network planning and simulation tools.