The energy transition with its change to renewable energy sources and a decarbonization of everyday life, presents system operators with significant challenges in their daily operation. At the same time, these changes also offer opportunities in the form of new sources of flexibilities. One prominent flexibility is the time-shifting of EV charging, which may be used to mitigate grid congestions. In Germany, this function is traditionally fulfilled through redispatch, a process that primarily targets generation-side adjustments. The central idea of redispatch 3.0 is to complement existing cost-based redispatch regimes (1.0 and 2.0) by means of voluntary and market-based demand-side flexibility.

In general, the decentralized flexibility of EVs is directly managed by flexibility service providers (FSP), who group EVs into flexibility pools, which allows for an improved forecast of the aggregated flexibility and power profiles. In a first step, FSPs register their flexibility pools, provide master data, and then may participate in a reoccurring market mechanism that contracts reserved capacity by matching the flexibility offers by FSPs with the predicted redispatch demand of TSOs for a given time window. After closure of the market, the winning FSPs deliver their predicted flexibility and power profiles to the TSOs on a regular basis. The TSO integrate this data into their regular redispatch processes, such as a clustering of multiple FSPs, as well as in the various forecasting processes. If an FSP should be activated, it receives an activation order, reoptimizes its power profiles based on the new restrictions and finally changes the charging of their EVs according to the new plan. To avoid creating new grid congestions on any grid level, the activation order may also contain further limitation for the replanning beyond the actual redispatch demand. The overall process ends with validation and balancing processes, which are done ex-post.

This approach has been implemented and evaluated in a real‑world redispatch project between the TSO TransnetBW and the FSP Octopus Energy. The project demonstrates the first end-to-end implementation of using aggregated household EV flexibility for redispatch, covering the full process chain from the TSO control room via the grid operator’s platform DA/RE to the FSP platform kraken to individual EV charging and subsequent imbalancing settlement. Over 700 EVs are already integrated under real‑world conditions, with the fleet size continuing to grow as a follow-up project is being continued, demonstrating operational scalability beyond laboratory or small‑scale pilots. Based on these 700 EVs, an average redispatch potential of approximately 2 MWh per day was offered to TransnetBW. The flexibility offers of Octopus Energy are provided after day-ahead market participation, while the activation decisions of TransnetBW are directly linked to the TSO’s established redispatch dimensioning procedures. Redispatch response quality proved high, with delivery rates consistently above 80%, comparable to the redispatch response quality of renewable resources like photovoltaics and wind power plants.

Summarizing, the goal of Redispatch 3.0 is to successfully integrate decentralized demand-side flexibility into the existing redispatch processes. The real-world project has shown promising results and proven the robustness of the newly developed concepts and ideas.