Application of artificial neural networks in design of lithium-ion batteries

Wu, Bin; Han, Sangwoo; Shin, Kang G.; Lu, Wei

doi:10.1016/j.jpowsour.2018.05.040

Cited by 91 publications

(55 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We observed several hurdles for the successful application of ML methods to battery research, where heterogeneous materials and properties are common. We describe the challenges of applying a machine learning approach to the discovery of battery materials [24][25][26][27][28].…”

Section: Introductionmentioning

confidence: 99%

Data-Driven Studies of Li-Ion-Battery Materials

2019

View full text Add to dashboard Cite

Batteries are a critical component of modern society. The growing demand for new battery materials—coupled with a historically long materials development time—highlights the need for advances in battery materials development. Understanding battery systems has been frustratingly slow for the materials science community. In particular, the discovery of more abundant battery materials has been difficult. In this paper, we describe how machine learning tools can be exploited to predict the properties of battery materials. In particular, we report the challenges associated with a data-driven investigation of battery systems. Using a dataset of cathode materials and various statistical models, we predicted the specific discharge capacity at 25 cycles. We discuss the present limitations of this approach and propose a paradigm shift in the materials research process that would better allow data-driven approaches to excel in aiding the discovery of battery materials.

show abstract

Section: Introductionmentioning

confidence: 99%

Data-Driven Studies of Li-Ion-Battery Materials

2019

View full text Add to dashboard Cite

show abstract

“…It is useful to evaluate the effect of each design parameter (e.g., the volume ratio of the active material particles, radius of the active material particles, pressure in the compaction process, and volume ratio of binder/additives) on the total specific resistance, as a guiding principle for manufacturing battery electrodes. One of the simplest approaches for this purpose is to evaluate the weight coefficients of the neurons on the first hidden layer of the ANN [29]. Here, we evaluate the summation of the weight coefficient magnitudes |w| of the first layer neurons for each input process parameter, as shown in Figure 8.…”

Section: Neural Network Regressionmentioning

confidence: 99%

Machine Learning Approaches for Designing Mesoscale Structure of Li-Ion Battery Electrodes

Takagishi¹,

Yamanaka²,

Yamaue³

2019

Batteries

View full text Add to dashboard Cite

We have proposed a data-driven approach for designing the mesoscale porous structures of Li-ion battery electrodes, using three-dimensional virtual structures and machine learning techniques. Over 2000 artificial 3D structures, assuming a positive electrode composed of randomly packed spheres as the active material particles, are generated, and the charge/discharge specific resistance has been evaluated using a simplified physico-chemical model. The specific resistance from Li diffusion in the active material particles (diffusion resistance), the transfer specific resistance of Li+ in the electrolyte (electrolyte resistance), and the reaction resistance on the interface between the active material and electrolyte are simulated, based on the mass balance of Li, Ohm’s law, and the linearized Butler–Volmer equation, respectively. Using these simulation results, regression models, using an artificial neural network (ANN), have been created in order to predict the charge/discharge specific resistance from porous structure features. In this study, porosity, active material particle size and volume fraction, pressure in the compaction process, electrolyte conductivity, and binder/additives volume fraction are adopted, as features associated with controllable process parameters for manufacturing the battery electrode. As a result, the predicted electrode specific resistance by the ANN regression model is in good agreement with the simulated values. Furthermore, sensitivity analyses and an optimization of the process parameters have been carried out. Although the proposed approach is based only on the simulation results, it could serve as a reference for the determination of process parameters in battery electrode manufacturing.

show abstract

“…More in precise, the model-based estimation with adaptive Kalman and particle filters or observers and fuzzy logic [40][41][42] or machine learning algorithms such as artificial neural networks (ANN) [43,44] and support vector machines (SVM) [45] are typically used for on-board implementation, taking into account their increased computational and memory requirements. On the other hand, the SoC estimation from Ah-counting [46] depends on the accuracy (sampling precision and frequency) of the current sensors and the initialization of the cell's capacity.…”

Section: In Discrete-time Domainmentioning

confidence: 99%

Comparative Study on Parameter Identification Methods for Dual-Polarization Lithium-Ion Equivalent Circuit Model

et al. 2019

View full text Add to dashboard Cite

Md.Sazzad.Hosen@vub.be (M.S.H.); Mohsen.Akbarzadeh.Sokkeh@vub.be (M.A.S.); Shovon.Goutam@vub.be (S.G); Joris.Jaguemont@vub.be (J.J.); Maitane.Berecibar@vub.be (M.B.); Joeri.van.mierlo@vub.be (J.V.M.) 2 Flanders Make,Abstract: A lithium-ion battery cell's electrochemical performance can be obtained through a series of standardized experiments, and the optimal operation and monitoring is performed when a model of the Li-ions is generated and adopted. With discrete-time parameter identification processes, the electrical circuit models (ECM) of the cells are derived. Over their wide range, the dual-polarization (DP) ECM is proposed to characterize two prismatic cells with different anode electrodes. In most of the studies on battery modeling, attention is paid to the accuracy comparison of the various ECMs, usually for a certain Li-ion, whereas the parameter identification methods of the ECMs are rarely compared. Hence in this work, three different approaches are performed for a certain temperature throughout the whole SoC range of the cells for two different load profiles, suitable for light-and heavy-duty electromotive applications. Analytical equations, least-square-based methods, and heuristic algorithms used for model parameterization are compared in terms of voltage accuracy, robustness, and computational time. The influence of the ECMs' parameter variation on the voltage root mean square error (RMSE) is assessed as well with impedance spectroscopy in terms of Ohmic, internal, and total resistance comparisons. Li-ion cells are thoroughly electrically characterized and the following conclusions are drawn: (1) All methods are suitable for the modeling, giving a good agreement with the experimental data with less than 3% max voltage relative error and 30 mV RMSE in most cases. (2) Particle swarm optimization (PSO) method is the best trade-off in terms of computational time, accuracy, and robustness. (3) Genetic algorithm (GA) lack of computational time compared to PSO and LS (4) The internal resistance behavior, investigated for the PSO, showed a positive correlation to the voltage error, depending on the chemistry and loading profile.

show abstract

Application of artificial neural networks in design of lithium-ion batteries

Cited by 91 publications

References 26 publications

Data-Driven Studies of Li-Ion-Battery Materials

Data-Driven Studies of Li-Ion-Battery Materials

Machine Learning Approaches for Designing Mesoscale Structure of Li-Ion Battery Electrodes

Comparative Study on Parameter Identification Methods for Dual-Polarization Lithium-Ion Equivalent Circuit Model

Contact Info

Product

Resources

About