(356d) Lithium-Ion Battery Model Development and Parameter Estimation for an Extended Single Particle Model Coupled with Thermal Modeling

Lithium-ion batteries are the predominant battery type utilized in portable consumer electronics and electric vehicles [1],[2],[3]. Commercially, batteries are improved based on the testing of the charge-discharge cycles. This testing generates parameters which are used in the battery management systems that monitor and control battery behaviours; but these procedures are time-consuming as they require multiple cycles to generate accurate and standardized parameters [3]. Therefore, it is vital to develop an accurate thermal model utilizing partial differential equations (PDEs) to capture the dynamics within the cell [1],[2]. In this work, a thermal model coupled with an electrochemical-based concentration model will be developed to indirectly determine ion-concentrations in the Lithium-ion cell and the developed model will be calibrated using nonlinear least squares optimization [3], as well as simulation/experimental data (temperature, conductivity, etc.) from testing on commercial Lithium-ion cells. The intent of this work is to develop a more accurate thermal model and couple the concentration and thermal models to determine the not directly measurable concentration properties in the cell, over the course of multiple charging/discharging cycles.

The primary difficulty in considering model type selection is the balance between computational efficiency and accuracy [3],[4],[5],[6]. Up-to-date modeling efforts utilize three major model types: the Doyle-Fuller-Neuman Model (DFN), the Single Particle Model (SPM). and Equivalent Circuit Models (ECM) [1]. While DFN models are the most accurate in portraying the electrochemical dynamics in the cell, they are also the most computationally intensive due to the characterization via PDEs coupled with nonlinear algebraic equations [3],[4],[5]. Furthermore, the increased complexity yields increased difficulty in parametrization of the model, which is an already challenging task [3],[4],[5]. The simplest models are ECMs, which seek to capture the battery dynamics using electrical components and circuitry [4]. While the fast computational time of these models is attractive for some applications, the downfall is the lack of representation of specific electrochemical dynamics in the cell; as such, ECMs fail to fully represent the internal physical characteristics which occur during charging/discharging cycles [4]. Due to the assumption that the electrodes may each be represented by a single spherical particle and by virtually eliminating the PDEs which deal with liquid phase diffusion, SPMs are much simpler than the DFN models, while also more accurate than ECM models and have a higher solving efficiency [2],[4],[5]. However, the cost of the simplification is in reduced accuracy; given the assumptions of the model, the electrolyte phase dynamics are not considered [4].

The motives of this work are to balance computational efficiency with accuracy by the use of an extended SPM (SPMe) which is coupled with a thermal model. The SPMe model uses the simplifying assumptions of the SPM, but also considers the liquid phase diffusion, so that the shortcomings of the SPM are addressed with only a slight decrease in computational efficiency [2],[4]. More specifically, this work proposes an extended SPM model coupled with a thermal model consisting of two ordinary differential equations (ODEs). The novelty of the model developed in this work is that very little literature currently exists on the coupling of SPMe and thermal models [2],[4]. Additionally, this work will consider a spatially varying and a non-spatially varying exchange current density both, as in Planella at al. [4] and Hwang et al. [6] respectively. As such, the effects of the associated assumptions will be carried into two different approaches to the voltage modeling, and the effect of these approaches on the electrolytic concentrations, voltage, and thermal model will be compared, similarly to work shown in [2],[4],[5],[6]. In this work, the developed model will be used to estimate parameters using nonlinear least squares optimization [3],[7],[8] and simulation/experimental data (temperature, conductivity, etc.) from testing on commercial Lithium-ion cells. The potential use of the developed model and parameters of this work into future control of monitoring efforts would allow for the consideration of thermal profiles without the time consuming and computationally challenging efforts of DFN or other Psuedo-2D model types, while also having more meaningful representation of the internal mechanisms than that of the ECM type models.

References

1. E. Prada, D. Di Domenico, Y. Creff, J. Bernard, V. Sauvant-Moynot, and F. Huet, “Simplified Electrochemical and Thermal Model of LiFePO 4 -Graphite Li-Ion Batteries for Fast Charge Applications,” J. Electrochem. Soc., vol. 159, no. 9, pp. A1508–A1519, 2012, doi: 10.1149/2.064209jes.
2. H. E. Perez, S. Dey, X. Hu, and S. J. Moura, “Optimal Charging of Li-Ion Batteries via a Single Particle Model with Electrolyte and Thermal Dynamics,” J. Electrochem. Soc., vol. 164, no. 7, p. A1679, Jun. 2017, doi: 10.1149/2.1301707jes.
3. Jin N, Danilov DL, Van den Hof PMJ, Donkers MCF. “Parameter estimation of an electrochemistry-based lithium-ion battery model using a two-step procedure and a parameter sensitivity analysis.” Int J Energy Res. 2018;42:2417–2430. https://doi.org/10.1002/er.4022.
4. F. B. Planella et al., “A continuum of physics-based lithium-ion battery models reviewed,” Prog. Energy, vol. 4, no. 4, p. 042003, Jul. 2022, doi: 10.1088/2516-1083/ac7d31.
5. G. Richardson, I. Korotkin, R. Ranom, M. Castle, and J. M. Foster, “Generalised single particle models for high-rate operation of graded lithium-ion electrodes: Systematic derivation and validation,” Electrochimica Acta, vol. 339, p. 135862, Apr. 2020, doi: 10.1016/j.electacta.2020.135862.
6. Gyuyeong Hwang, Niranjan Sitapure, Jiyoung Moon, Hyeonggeon Lee, Sungwon Hwang, Joseph Sang-Il Kwon, “Model predictive control of Lithium-ion batteries: Development of optimal charging profile for reduced intracycle capacity fade using an enhanced single particle model (SPM) with first-principled chemical/mechanical degradation mechanisms,” Chemical Engineering Journal, Volume 435, Part 1, 2022, 134768, ISSN 1385-8947, https://doi.org/10.1016/j.cej.2022.134768
7. S. Santhanagopalan, Q. Guo, and R. White, “Parameter Estimation and Model Discrimination for a Lithium-Ion Cell,” Journal of the Electrochemical Society, vol. 154, no. 3, pp. A198–A206, Jan. 2007.
8. H. Dan, N. Yamashita, and M. Fukushima, “Convergence properties of the inexact Levenberg-Marquardt method under Local Error Bound Conditions,” Optimization Methods and Software, vol. 17, no. 4, pp. 605–626, Oct. 2010.