A Physics-Informed Machine Learning Approach for Estimating Lithium-Ion Battery Temperature

Cho, Gyouho; Wang, Mengqi; Kim, Youngki; Kwon, Jaerock; Su, Wencong

doi:10.1109/access.2022.3199652

Cited by 23 publications

(10 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…108 Researchers have still devoted themselves to using ML to forecast the battery's internal thermal behavior. Up till now, multi-features such as cell external temperature, depth of discharge, nominal capacity, ambient temperature, and discharge rate have been utilized to train the ML, realizing a prediction of thermal effects (generally expressed as heat generation rate 109,110 /internal temperature 111 /external temperature 112,113 ).…”

Section: Thermal-based Tasksmentioning

confidence: 99%

Feature–target pairing in machine learning for battery health diagnosis and prognosis: A critical review

Huang

Sugiarto

2023

EcoMat

View full text Add to dashboard Cite

Lithium‐ion batteries (LIBs) have been dominating the markets of electric vehicles and grid energy storage. Accurate monitoring of battery health status has been one of the most critical challenges of the battery industry. Machine learning (ML) has been widely applied to battery health estimation as well as prediction. Here, by investigating the specific features and targets, we comprehensively discuss task‐oriented ML implementation in various application scenarios in the field of battery health. This review explores the tasks assisted by ML based on multi‐level cell degradation. We highlight opportunities and significance of considering the potential feature–target pair during the ML model training to identify more health information about LIBs as well as shed light into designing tasks for new application scenarios.

show abstract

Section: Thermal-based Tasksmentioning

confidence: 99%

Feature–target pairing in machine learning for battery health diagnosis and prognosis: A critical review

Huang

Sugiarto

2023

EcoMat

View full text Add to dashboard Cite

show abstract

“…Here, the authors used FVM to numerically solve the electrochemical-thermal coupled model, which is used to generate the training data for an LSTM network. In the work of [32], the lumped capacitance thermal equation is used in the loss function of the PINN to predict the temperature of LIB cells. Here, the network is trained with battery test data, and the heat equation is used to identify the thermal behavior of the cell.…”

Section: State Of the Research: Hybrid Modeling Of Lithium-ion Batteriesmentioning

confidence: 99%

Hybrid Modeling of Lithium-Ion Battery: Physics-Informed Neural Network for Battery State Estimation

Singh,

Ebongue,

Rezaei

et al. 2023

Batteries

View full text Add to dashboard Cite

Accurate forecasting of the lifetime and degradation mechanisms of lithium-ion batteries is crucial for their optimization, management, and safety while preventing latent failures. However, the typical state estimations are challenging due to complex and dynamic cell parameters and wide variations in usage conditions. Physics-based models need a tradeoff between accuracy and complexity due to vast parameter requirements, while machine-learning models require large training datasets and may fail when generalized to unseen scenarios. To address this issue, this paper aims to integrate the physics-based battery model and the machine learning model to leverage their respective strengths. This is achieved by applying the deep learning framework called physics-informed neural networks (PINN) to electrochemical battery modeling. The state of charge and state of health of lithium-ion cells are predicted by integrating the partial differential equation of Fick’s law of diffusion from a single particle model into the neural network training process. The results indicate that PINN can estimate the state of charge with a root mean square error in the range of 0.014% to 0.2%, while the state of health has a range of 1.1% to 2.3%, even with limited training data. Compared to conventional approaches, PINN is less complex while still incorporating the laws of physics into the training process, resulting in adequate predictions, even for unseen situations.

show abstract

“…For instance, the pre-layer and connection layer structures originated from the analytical solution of the physic law improved the prediction outcomes of the PINN in the literature published by Zobeiry et al [27]. This study implements the pre-layer and connection layer architectures between the input and hidden layers with the sine and exponential activation functions as proposed in [27] - [28]. Two architectures with different input sets are prepared.…”

Section: Pre-layer Architecturementioning

confidence: 99%

An LSTM-PINN Hybrid Method to Estimate Lithium-Ion Battery Pack Temperature

et al. 2022

Self Cite

View full text Add to dashboard Cite

Physics-based models for battery temperature prediction are often not suitable for online applications due to the large number of fitted parameters, low fidelity results from parameter inaccuracy and unaccounted model dynamics, limited high quality experimental data, and slow convergence of predictions from unknown initial conditions. On the other hand, data-driven models require much less computation power but a large dataset to learn the time dependent behavior. This paper proposes a physics-informed neural network (PINN) to take full advantage of both physics-based and data-driven models. Four comparative studies were performed to investigate the effectiveness of including chamber temperature with two different activation functions, the optimal number of neurons and hidden layers, the dependance on reversible heat generation, and the performance of long short-term memory (LSTM)-PINN model. The results show that the LSTM-PINN with chamber temperature as one of the inputs delivers better prediction accuracy. The LSTM-PINN with an exponential activation function for the chamber temperature has a more accurate prediction for the direct current fast charge (DCFC) test profile and similar prediction accuracy for the grade load (GL) 100 test profile. The root mean square errors (RMSEs) of the LSTM-PINN are 0.57°C for DCFC and 0.52°C for GL 100, respectively. In addition, having 55 neurons and 4 hidden layers gives the lowest prediction error. Furthermore, the improvement by having reversible heat generation is negligible. At last, the LSTM-PINN model has less prediction error than the LSTM model, especially when the battery temperature range is enormous.

show abstract

A Physics-Informed Machine Learning Approach for Estimating Lithium-Ion Battery Temperature

Cited by 23 publications

References 44 publications

Feature–target pairing in machine learning for battery health diagnosis and prognosis: A critical review

Feature–target pairing in machine learning for battery health diagnosis and prognosis: A critical review

Hybrid Modeling of Lithium-Ion Battery: Physics-Informed Neural Network for Battery State Estimation

An LSTM-PINN Hybrid Method to Estimate Lithium-Ion Battery Pack Temperature

Contact Info

Product

Resources

About