16 January 2025 in IEEE Transactions on Transportation Electrification
A key challenge that is currently hindering the widespread use of retired Electric Vehicle (EV) batteries for Second-Life (SL) applications is the ability to accurately estimate and monitor their State of Health (SOH). Second-life battery systems can be sourced from different battery packs with a lack of knowledge of their historical usage. Accurate SOH estimation is critical because it enables reliable performance, safety, and optimal utilization of SL batteries, ensuring they meet the requirements of various applications including grid energy storage and backup power. In this work, for in-the-field use of SL batteries, we introduce an online adaptive health estimation approach with the guarantees of Bounded-Input-Bounded-Output (BIBO) stability. This method relies exclusively on operational data that can be accessed in real time from SL batteries. The effectiveness of the proposed approach is shown on a laboratory-aged experimental data set of retired EV batteries.
21 October 2024 in Data in Brief
This paper describes an experimental dataset of used lithium-ion battery cells cycled on grid storage synthetic duty cycles to study their feasibility for second-life applications. Data were collected at the Stanford Energy Control Laboratory at Stanford University, CA, USA. The ten INR21700-M50T battery cells with graphite/silicon anode and Nickel–Manganese–Cobalt (NMC) cathode had been previously tested over a period of 23 months according to the Urban Dynamometer Driving Schedule (UDDS) discharge driving profile. In this paper, six out of these ten cells are tested for a period of 24 months. The aging campaign is a combination of calendar aging and cycling. The cycling portion is designed to replicate real-world usage patterns based on synthetic duty cycles for residential and commercial grid energy storage systems (ESS). Battery cycling alternates between 20 °C and 35 °C to simulate seasonal temperature variations encountered in grid applications. The calendar aging, on the other hand, occurs at room temperature. Periodic assessments of battery degradation during second-life testing are accomplished via Reference Performance Tests for second-life (RPT S) comprising of a combined capacity and pulse power test, and Electrochemical Impedance Spectroscopy (EIS) at three state-of-charge (SOC) values. The data set captures the combined effects of cycling-induced stress and long-term storage.
19 November 2025 in Data in Brief
This paper presents a comprehensive dataset of 22 NMC/graphite lithium-ion batteries subjected to both calendar and cycle aging under controlled operating conditions. The cells were aged at 0 °C, 25 °C, and 40 °C for a duration of 90 days. Periodic reference performance tests (RPTs) were conducted throughout the aging process. Each RPT included a capacity test and an electrochemical impedance spectroscopy (EIS), both performed at 25 °C. EIS was repeated at state-of-charge (SOC) values of 0 %, 25 %, 50 %, 75 % and 100 %. Cycle-aged cells were operated between various voltage limits at two distinct charging rates, while calendar-aged cells were stored at either 80 % or 100 % state-of-charge (SOC). This dataset captures a diverse range of degradation pathways. It can be used to develop and validate battery aging models, state-of-health (SOH) estimation algorithms, and support the design of predictive maintenance strategies for lithium-ion battery systems.
9 April 2026 in SSRN Electronic Journal
Electrochemical impedance spectroscopy (EIS) is a non-invasive technique for probing internal battery processes, but full-spectrum measurements are impractical for onboard battery management systems (BMSs). This paper presents a constrained optimization framework to identify a minimal, health-informative set of EIS excitation frequencies from battery data, enabling partial EIS measurements suitable for real-time BMS deployment. Frequency selection is formulated as a nonconvex optimization problem solved via particle swarm optimization (PSO), with a log-determinant objective that promotes a well-conditioned distribution of relaxation times (DRT) kernel and a Spearman rank correlation constraint that ties the selected frequencies to capacity fade. Initialization sensitivity is mitigated through K-means clustering across multiple PSO restarts. A key finding is that as few as three optimally selected frequencies yield a well-conditioned DRT kernel invertible without regularization, eliminating the computationally intensive regularization parameter selection of conventional DRT methods. Features extracted from the reduced DRT curves are used for data-driven state-of-health (SOH) estimation. Offline validation on two nickel manganese cobalt (NMC) datasets with distinct anode compositions yields mean absolute percentage errors (MAPEs) of 1.49% and 1.09% for a data-driven model, demonstrating accuracy and generalizability across cell chemistries.
17 September 2026 in Journal of Power Sources