This paper presents a simulation-based analysis of a Hybrid Battery Energy Storage System (HBESS) that combines a commercial nickel manganese cobalt (NMC811) cell with a next-generation cell featuring Silicon-Graphite (SiGr) and lithium nickel manganese oxide (LNMO) electrodes. Using two Pseudo Two-Dimensional (P2D) physics-based models, one for each cell in the HBESS, the system’s performance and degradation are evaluated. The work begins by outlining the current Lithium-Ion Battery (LIB) landscape and emphasizing emerging chemistries. The methodology used to calibrate the P2D model of the LNMO–SiGr cell is also described and validated against experimental full-cell data. A real-world stationary energy storage scenario is simulated to assess the benefits of hybridization. The proposed HBESS architecture reduces the cobalt content of the battery and mitigates degradation risks associated with silicon expansion by operating the LNMO–SiGr cell within a controlled 80–20% State of Charge (SOC) window. The NMC811 cell manages the system’s base load, ensuring stability, while the LNMO–SiGr cell provides supplementary energy during peak demands. This hybrid approach has the potential to enhance system durability, sustainability, and overall performance by leveraging the strengths of different lithium-ion chemistries.

This conference paper explores hybrid battery energy storage systems (HBESS) through a simulation-based analysis, combining a commercial nickel manganese cobalt oxide cell with a next-generation cell and evaluating their performance under a realistic application scenario. The study employs physics-based models (PBMs) to simulate and estimate system performance. The paper begins with an overview of actual lithium-ion battery technologies, providing context on the research landscape, with a focus on cobalt-free and graphite-silicon electrodes. It then outlines the PBM used, followed by the parameterization of the cells forming the HBESS. A detailed description is provided of the steps undertaken to calibrate the next-generation cell model, which has been validated using experimental full-cell data. The article further discusses key aspects of the hybrid system, including the application of model predictive control strategies and critical operational considerations. A case scenario, based on an electric vehicle usage pattern, is then evaluated through simulation to assess the potential benefits in terms of performance optimization and degradation mitigation. In this way, the study demonstrates how hybridization strategies can leverage the strengths of different battery technologies when integrated with actual battery applications, reducing degradation and cobalt content in battery packs.

Accurate half-cell open-circuit potential functions are essential for physics-based battery models but are traditionally obtained through invasive and time-consuming procedures. This work introduces a systematic procedure that reconstructs half-cell potentials from full-cell voltage data, differential capacity analysis, and prior knowledge of stoichiometrically constrained open-circuit potential trust regions, eliminating the need for cell teardown. Validation against reference values demonstrates high accuracy, offering a non-invasive and scalable solution to accelerate the adoption of advanced electrochemical models to enhance battery performance.

This conference paper presents a comparative study of various models used to characterize the behavior of NMC Li-ion batteries. Specifically, this work focuses on analyzing data granularity and examining important features such as model accuracy and required computing resources. The paper presents the aforementioned study employing two different types of models: a physics-based model found in the literature and the well-known equivalent circuit model. Each of these models are formulated and calibrated using different techniques based on their typology. To conduct these studies, a real NMC811 lithium-ion battery with a 5Ah capacity is considered. A set of charging and discharging profiles were used to analyze and compare the results obtained from different models.