Curative grid operator intervention by using peak smoothing to avoid grid overloads

The future integration of battery-powered electric vehicles is a challenge for low-voltage grids in particular. With high penetration of electric vehicles in the grid, overloads and voltage band violations can occur due to the resulting high simultaneous load in the grid, depending on the charging strategies, the plug-in behaviour and other factors. In addition to grid expansion, grid congestion can also be avoided through the use of grid-serving flexibility, such as the postponement and balancing of simultaneous charging processes. In this article a simulation result prepared as a video is used to illustrate how overloads of the substation transformer and voltage band violations in a low-voltage grid can be eliminated by reducing or postponing charging processes of electric vehicles.

The simulation was carried out as part of the project Bidirectional Charging Management (BCM) / Bidirektionales Lademanagement (BDL) using the FfE’s distribution grid and energy system model, GridSim. The implementation in the simulation is based on the “peak smoothing” model, a proposed extension of § 14a EnWG, from the BET/EY study “Regulierung, Flexibilisierung und Sektorkopplung” commissioned by the German Federal Ministry for Economic Affairs and Energy (BMWi) and published in 2018 [1]. A detailed description of the assumptions and results of the simulation was presented at the sixth E-Mobility Power System Integration Symposium under the title “AVOIDING LOW-VOLTAGE GRID OVERLOADS THROUGH CURATIVE GRID OPERATOR INTERVENTION WITH FOCUS ON ELECTRIC VEHICLES”.

In the following video (with german audio) the curtailment of charging processes to avoid grid congestion is illustrated with an animated map of a low-voltage grid:


By loading the video, you agree to YouTube's privacy policy.
Learn more

Load video

The following key parameters, assumptions and limitations, among others, apply to the simulation results presented:

  • The grid shown is a rural or suburban grid with households and businesses. It supplies 66 grid connection points (GCPs). The scenario refers to the year 2040 and includes a high penetration of heat pumps and electric vehicles (EV). 67 % of the GCPs cover their heat demand with a heat pump. 30 % of the GCPs have a photovoltaic (PV) system. 55 % of the GCPs have at least one EV assigned to them, 55 in total, up to four per GCP.
  • 49 % of the 55 EVs are bidirectional and feed back into the buildings or the grid. 33 % are charging in a tarrif-optimised manner and participate in electricity trading on the spot market (vehicle-to-grid with arbitrage trading). These are located at 12 % of the GCPs. Especially with market-oriented use of flexibility such as tariff-optimised charging or arbitrage trading on the spot market, grid congestion can occur due to high charging simultaneity, as shown in the paper for the conference “Zukünftige Stromnetze”. 16 % of the EVs charge optimised on the surplus from the PV system at their GCP and feed back into the building (vehicle-to-home with PV self-consumption optimisation).
  • For the simulation, it is assumed that the grid operator has almost real-time knowledge of all voltages and currents in the grid via measurement data. The load of the substation transformer and the voltage at its busbars is measured. The load and voltage at all GCPs are monitored via the smart meter infrastructure. Using the load of the GCPs, the utilisation of the lines is estimated simulatively with a digitalised grid model.
  • Using the smart meters and additional control components, the grid operator can gradually reduce the load of charging electric vehicles if necessary. In line with the current § 14a EnWG, this reduction may not exceed two hours per customer and day in the case of complete reduction. However, the power can also be reduced to 60 % and 30 %. The curtailed time is weighted with the amount of curtailment, i.e. if the charging power of a customer is only reduced to 60 % on one day, the maximum permitted duration of intervention increases to five hours.
  • In the simulation, a voltage drop of 6 % below 94 % of the nominal voltage at a grid interconnection point is already considered a voltage band violation, not 90 % as in reality, since voltage fluctuations in the superimposed medium-voltage grid are not modelled in the simulation.
  • If 100 % of the nominal current is exceeded in one of the windings of the transformer or one phase of a three-phase cable, these are considered overloaded in the simulation.
  • In the current implementation, regardless of the type and location of the congestion, all electric vehicles are gradually reduced in their power until the congestion is eliminated. In future, the simulation should differentiate more finely in the case of voltage band violations and line overloads to ensure that only those charging processes are reduced that have a noticeable influence on the overloads depending on their spatial position in the grid.

In the example shown in the video, it is possible to prevent grid overloads caused by market-optimised electric vehicles by monitoring the grid and reducing the charging power as needed. In our paper “AVOIDING LOW-VOLTAGE GRID OVERLOADS THROUGH CURATIVE GRID OPERATOR INTERVENTION WITH FOCUS ON ELECTRIC VEHICLES”, the effects of peak smoothing are statistically evaluated for 1206 low-voltage grids and the following questions are examined:

  • To what extent can grid expansion be avoided through the use of peak smoothing?
  • To what extent are drivers of electric vehicles disadvantaged by lower battery levels?
  • To what extent does the load and energy balance of the grid areas change?


[1] Zander, Wolfgang et al.: Digitalisierung der Energiewende – Topthema 2: Regulierung, Flexibilisierung und Sektorkopplung. Berlin: Bundesministerium für Wirtschaft und Energie (BMWi), 2018.

More Information: