Can an identifiability-optimizing test protocol improve the robustness of subsequent health-conscious lithium-ion battery control? an illustrative case study

Liu, Ji; Rothenberger, Michael J.; Mendoza, S.; Mishra, Partha; Jung, Yun-sik; Fathy, Hosam K.

doi:10.1109/acc.2016.7526663

Cited by 14 publications

(10 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…12,13 However, solving a large-scale optimal control problem with thousands of states and nonlinear dynamics given by[2]-[4],[7]-[9] is computationally intractable. Two weeks of wall clock time are needed to optimize a 100 second input trajectory for the electrochemical model on a PC with an INTEL Core i5 -1.8GHz dual core, Turbo Boost up to 2.9GHz, with 3MB shared L3 cache, and 8GB RAM.…”

mentioning

confidence: 99%

Optimal Experimental Design for Parameterization of an Electrochemical Lithium-Ion Battery Model

Park

Kato

Gima

et al. 2018

J. Electrochem. Soc.

100

View full text Add to dashboard Cite

We consider the problem of optimally designing an excitation input for parameter identification of an electrochemical Li-ion battery model. The conventional approach to performing parameter identification uses standard test cycles. In contrast, we optimally design the input trajectory to maximize parameter identifiability in the sense of Fisher information. Specifically, we derive sensitivity equations for the electrochemical model. This approach enables parameter sensitivity analysis and optimal parameter fitting via gradient-based algorithms. This paper presents a general systematic approach to identify the electrochemical parameters in a non-invasive way. First, we group parameters into two sets: (i) equilibrium parameters, and (ii) dynamical parameters. We also divide the dynamical parameters into subsets by calculating orthogonalized sensitivity, which mitigates linear dependence between parameters. A large number of input profiles have been devised to constitute an input library. Then, the optimal inputs are selected from the input library to maximize the Fisher information, via convex programming. Using this framework a number of relevant experiments are obtained to parameterize. To validate our approach experimentally, we consider a 18650 Lithium nickel cobalt aluminum oxide battery. Compared to the conventional approach, our proposal achieves lower voltage RMSE across all experimental testing cycles.

show abstract

mentioning

confidence: 99%

Optimal Experimental Design for Parameterization of an Electrochemical Lithium-Ion Battery Model

Park

Kato

Gima

et al. 2018

J. Electrochem. Soc.

100

View full text Add to dashboard Cite

show abstract

“…The state and input variables can then be recovered using the flat output trajectory according to Eq. (28). Therefore, the application of differential flatness makes it possible to represent battery dynamics in a computationally efficient manner.…”

Section: B Application To Batteriesmentioning

confidence: 99%

“…The volume fraction of the active material ε j in Eq. (4) is one of the SOH parameters relating to electrode capacity in SPM and hence cell capacity, which changes with aging for a given battery [28]. Suppose the controller assumes each electrode has its original capacity, while the actual capacity is only 90% of the original due to aging.…”

Section: B Traditional CCCV Chargingmentioning

confidence: 99%

An Extended Differential Flatness Approach for the Health-Conscious Nonlinear Model Predictive Control of Lithium-Ion Batteries

Liu

Fathy

2017

IEEE Trans. Contr. Syst. Technol.

Self Cite

View full text Add to dashboard Cite

This article examines the problem of optimizing lithium-ion battery management online, in a health-conscious manner. This is a computationally intensive problem. Previous work by the authors addresses this challenge by exploiting the differential flatness of Fick's law of diffusion to improve computational efficiency, but is limited by the fact that the dynamics of a full battery cell are not differentially flat, even when the individual battery electrode dynamics are. The article addresses this challenge by extending the application of differential flatness to a full single particle model (SPM). Specifically, we use the conservation of charge to express the flat output trajectory of one electrode as an affine function of the other electrode's flat output trajectory. In this way, we enforce differential flatness for the full battery model. This makes it possible to express the battery charge/discharge trajectory in terms of one flat output trajectory. We optimize this trajectory using a pseudospectral method. This reduces the computational cost of the optimization by about a factor of 5 compared to pseudospectral optimization alone. Additionally, the robustness of the nonlinear model predictive control (NMPC) strategy is demonstrated in simulation by adding state of health (SOH) parameter uncertainties.

show abstract

“…The Design of Experiment (DoE) consists in finding the optimal input sequence able to minimize the uncertainty of model parameters. 23 In Liu et al 24 the importance of DoE in the context of control of lithium-ion batteries has been underlined. This topic has been discussed by many authors.…”

Section: Introductionmentioning

confidence: 99%

“…The design of experiment (DoE) consists in finding the optimal input sequence able to minimize the uncertainty of model parameters. 22 In Liu et al 23 the importance of DoE in the context of control of lithium-ion batteries has been underlined. In Mendoza et al 24 the thermal cycle of a cell has been optimized in order to estimate the entropy coefficient while reducing experimental time and maximizing the Fisher matrix.…”

Section: Introductionmentioning

confidence: 99%

Optimal Design of Experiments for a Lithium-Ion Cell: Parameters Identification of an Isothermal Single Particle Model with Electrolyte Dynamics

Pozzi

Ciaramella

Volkwein

et al. 2018

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Advanced battery management systems rely on mathematical models to guarantee optimal functioning of Lithium-ion batteries. The Pseudo-Two Dimensional (P2D) model is a very detailed electrochemical model suitable for simulations. On the other side, its complexity prevents its usage in control and state estimation. Therefore, it is more appropriate the use of simplified electrochemical models such as the Single Particle Model with electrolyte dynamics (SPMe), which exhibits good adherence to real data when suitably calibrated. This work focuses on a Fisher-based optimal experimental design for identifying the SPMe parameters. The proposed approach relies on a nonlinear optimization to minimize the covariance parameters matrix.At first, the parameters are estimated by considering the SPMe as the real plant.Subsequently, a more realistic scenario is considered where the P2D model is used 1 arXiv:1811.08656v2 [cs.SY] 30 Sep 2019 to reproduce a real battery behavior. Results show the effectiveness of the optimal experimental design when compared to standard strategies.

show abstract

Can an identifiability-optimizing test protocol improve the robustness of subsequent health-conscious lithium-ion battery control? an illustrative case study

Cited by 14 publications

References 22 publications

Optimal Experimental Design for Parameterization of an Electrochemical Lithium-Ion Battery Model

Optimal Experimental Design for Parameterization of an Electrochemical Lithium-Ion Battery Model

An Extended Differential Flatness Approach for the Health-Conscious Nonlinear Model Predictive Control of Lithium-Ion Batteries

Optimal Design of Experiments for a Lithium-Ion Cell: Parameters Identification of an Isothermal Single Particle Model with Electrolyte Dynamics

Contact Info

Product

Resources

About