A Fast, Memory-Efficient Discrete-Time Realization Algorithm for Reduced-Order Li-Ion Battery Models

Gopalakrishnan, Krishnakumar; Zhang, Teng; Offer, Gregory J.

doi:10.1115/1.4035526

Cited by 10 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Nonetheless, for high performance applications such as in automotive BMSs there is an overwhelming demand for obtaining insight into internal cell variables. In view of this consideration, model order reduction techniques play a key role in porting a relevant subset of the P2D model onto a real-time microcontroller [13]- [16]. Owing to its simplicity, the Single Particle Model (SPM) is a promising Reduced Order Model (ROM) candidate with the potential to usher in the era of PBMs in embedded applications, and is summarized in section I-B.…”

Section: A the Doyle-fuller-newman (Dfn) Modelmentioning

confidence: 99%

A Composite Single Particle Lithium-Ion Battery Model Through System Identification

Gopalakrishnan

Offer

2022

IEEE Trans. Contr. Syst. Technol.

Self Cite

View full text Add to dashboard Cite

In the realm of Lithium-ion (Li-ion) battery modeling, owing to its simplicity, the Single Particle Model (SPM) has long been considered to be a promising Reduced Order Model (ROM) candidate to usher in the era of Physics-Inspired Models (PIMs) in embedded applications. However, at high load currents, the standard SPM exhibits poor accuracies in computing the cell's terminal voltage, thereby rendering it unsuitable as the plant model in state-estimation tasks. A comprehensive evaluation of the salient electrolyte-enhanced SPMs from literature reveals that current solutions are either mathematically intractable or overly simplistic. For the ionic concentration in the electrolyte, the well-known quadratic approximation model, which straddles the boundary of computational complexity and mathematical tractability, reveals a poor temporal performance, particularly at the current collector interfaces. In this work, we retain the spatial dynamics of the quadratic approximation model, whilst proposing a novel approach using system identification techniques for its temporal dynamics. By employing linear approximations for the relevant subsystems, we identify discrete-time transfer functions of the de-biased number of moles per unit area of lithium ions in the electrolyte within each electrode region, yielding improved spatiotemporal accuracies for the electrolyte concentration profile. We then augment the standard SPM with the new system identificationbased electrolyte dynamics to arrive at an Electrolyte-Enhanced Composite Single Particle Model (EECSPM). Finally, we demonstrate superior performance of the EECSPM when compared with the incumbent state-of-the-art, thereby representing a concrete advancement towards the goal of using PIMs in real-time applications.

show abstract

Section: A the Doyle-fuller-newman (Dfn) Modelmentioning

confidence: 99%

A Composite Single Particle Lithium-Ion Battery Model Through System Identification

Gopalakrishnan

Offer

2022

IEEE Trans. Contr. Syst. Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…This is seen as a widely available cell that provides a strong reference for the current state of a high energy conventional intercalation cell with an NCM 811 positive electrode and bi-component Graphite-SiO x negative electrode. To the authors knowledge, the discrete realisation algorithm (DRA) has been validated from full-order and linearised partial differential implementations [36,20,49,50] ; however, an experimental validation has not been presented in literature. In this section, an experimental voltage validation of the CI-DRA utilising LiiBRA.jl and the parameterised LG Chem.…”

Section: Experimental Validationmentioning

confidence: 99%

A Computationally Informed Realisation Algorithm for Lithium-Ion Batteries Implemented with LiiBRA.jl

Planden¹,

Lukow²,

Henshall³

et al. 2022

Preprint

View full text Add to dashboard Cite

Real-time battery modelling advancements have quickly become a requirement as the adoption of battery electric vehicles (BEVs) has rapidly increased. In this paper an open-source, improved discrete realisation algorithm, implemented in Julia for creation and simulation of reduced-order, real-time capable physics-based models is presented. This work reduces the Doyle-Fuller-Newman electrochemical model into continuousform transfer functions and introduces a computationally informed discrete realisation algorithm (CI-DRA) to generate the reduced-order models. Further improvements in conventional offline model creation are obtained as well as achieving in-vehicle capable model creation for ARM based computing architectures. Furthermore, a sensitivity analysis on the resultant computational time is completed as well as experimental validation of a worldwide harmonised light vehicle test procedure (WLTP) for a LG Chem. M50 21700 parameterisation. A performance comparison to the conventional Matlab implemented discrete realisation algorithm (DRA) is completed showcasing a mean computational time improvement of 88%. Finally, an ARM based compilation is investigated for in-vehicle model generation and shows a modest performance reduction of 43% when compared to the x86 implementation while still generating accurate models within 5.5 seconds.

show abstract

“…However, numerical models require strong computation power and are therefore less suitable to be integrated into BMS applications for temperature estimation [133]. For that reason, computationally intensive models can be reduced to simplified numerical models that are less accurate but significantly reduce computations times [84,134,135]. Alternatively, EQC models can be used, which have simple constructions and enable fast calculations.…”

Section: Heat Balancementioning

confidence: 99%

A review on various temperature-indication methods for Li-ion batteries

et al. 2019

View full text Add to dashboard Cite

Temperature measurements of Li-ion batteries are important for assisting Battery Management Systems in controlling highly relevant states, such as State-of-Charge and State-of-Health. In addition, temperature measurements are essential to prevent dangerous situations and to maximize the performance and cycle life of batteries. However, due to thermal gradients, which might quickly develop during operation, fast and accurate temperature measurements can be rather challenging. For a proper selection of the temperature measurement method, aspects such as measurement range, accuracy, resolution, and costs of the method are important. After providing a brief overview of the working principle of Li-ion batteries, including the heat generation principles and possible consequences, this review gives a comprehensive overview of various temperature measurement methods that can be used for temperature indication of Li-ion batteries. At present, traditional temperature measurement methods, such as thermistors and thermocouples, are extensively used. Several recently introduced methods, such as impedance-based temperature indication and fiber Bragg-grating techniques, are under investigation in order to determine if those are suitable for largescale introduction in sophisticated battery-powered applications.

show abstract

A Fast, Memory-Efficient Discrete-Time Realization Algorithm for Reduced-Order Li-Ion Battery Models

Cited by 10 publications

References 24 publications

A Composite Single Particle Lithium-Ion Battery Model Through System Identification

A Composite Single Particle Lithium-Ion Battery Model Through System Identification

A Computationally Informed Realisation Algorithm for Lithium-Ion Batteries Implemented with LiiBRA.jl

A review on various temperature-indication methods for Li-ion batteries

Contact Info

Product

Resources

About