Simplified modeling approach for predicting the remaining useful life of EV batteries in second-life applications

An accurate assessment of the Remaining Useful Life (RUL) of battery modules is crucial for the technical and economic feasibility of repurposing electric vehicle (EV) batteries for second-life applications. This assessment requires, two key components: determining the battery’s health at the end of its first life cycle, and predicting its aging behavior during second-life use. However, because battery aging behavior is highly influenced by electrode chemistry, manufacturing, and usage conditions, making aging prognosis is a complex task. Although various modeling methods exist, semi-empirical aging models appears as the most effective even if require extensive experimental data for parameter calibration. This data is usually obtained through prolonged testing under various specific conditions that depend on the severity of the aging process. Over the past decade, multiple generations of EV batteries have been introduced to the market. Traditional modeling approaches for each brand of battery would require substantial and costly experimental data. To address this challenge, this paper proposes a simplified modeling approach to characterize aging behavior under deliberately restricted operating conditions. A four-parameter model was developed to describe the evolution of calendar and cycling aging in battery modules considering cumulative aging mechanisms effect without path dependencies. The model parameters were calibrated using experimental data from tests conducted on modules from two retired 23 kWh Renault Zoe battery packs from electric vehicles (EVs) with different states of health (SoH). The model was then used to calculate an expected lifespan of the batteries’ modules for different second-life applications.