Profile Generation For Ancillary Service Provision Using Repurposed Electric Vehicle Batteries In Stationary Energy Storage Systems

The number of electric vehicles (EVs) has been increasing. According to estimates, by 2030, the proliferation of EVs will result in the availability of 100–200 GWh/year of electricity storage through EVs.

After being used in EVs under highly dynamic charging and discharging profiles and reaching a particular capacity level (e.g., 80% of the beginning of life (BoL) capacity), EV battery cells can be repurposed, in their second life, for less-demanding applications such as stationary energy storage systems (ESSs).

To evaluate the viability and assess degradation of batteries in their second life while providing ancillary services, this paper proposes a two-stage framework for generating second-life cycling profiles that mimic battery energy storage system (BESS) operation. Since these cycles differ significantly from typical driving cycles, we use energy throughput as a reference metric, which can be translated into an equivalent number of cycles. The proposed profile generation method is generic and applies to different types of ancillary services, battery sizes, and chemistries.

We use a generic per-unit power signal that represents the requested power from a BESS for a selected service (e.g., frequency containment reserve, self-consumption). This signal can be derived from available historical data. In the first stage, the signal is scaled to adapt to a desired battery size. We maintain the state-of-energy (SOE) within a predefined range by optimally offsetting the power profile. This is achieved by formulating and solving an optimization problem in terms of battery power and state-of-energy (SOE). The resulting optimal power profile serves as the input for the second stage.

In the second stage, we verify whether the battery’s operational constraints—such as current and voltage limits—are satisfied. This is done by converting the power profile into a current profile using a BESS equivalent circuit model (ECM). The modeller can use any generic ECM. If any constraints are violated, the algorithm iteratively adjusts the first-stage decisions—specifically, the scaling of the signal and/or the power offset—and re-evaluates the constraints. This process repeats until all operational limitations are met.

The final profile is first validated numerically as part of the second stage, and subsequently tested experimentally on a real BESS module built from repurposed EV cells. This validation is essential to ensure safe operation during second-life battery testing.