What question did this study set out to answer?

This research aims to enhance the accuracy of voltage predictions for lithium-ion batteries using a hybrid modeling approach.

February 8, 2026Energy StorageOpen Access

Hybrid Physics–Informed and Machine Learning Model for Accurate Lithium‐Ion Battery Voltage Prediction

View Full Paper

Ask AI

Bookmark

Authors

Hybrid model improves lithium-ion battery voltage prediction accuracy, highlighting implications for energy storage systems.

This research aims to enhance the accuracy of voltage predictions for lithium-ion batteries using a hybrid modeling approach.
Integrates an equivalent circuit model (ECM) with a multilayer perceptron (MLP) neural network.
Utilizes the Red-Tailed Hawk optimization algorithm to determine ECM parameters.
Evaluates the model using experimental data from a commercial lithium-ion battery under dynamic current profiles.
Conducts an ablation study to assess the impact of network depth on accuracy.
Compares performance against decision tree and random forest algorithms.
Achieves a reduction in prediction error, with RMSE decreasing from 0.1521 V to 66.6 mV.
Mean absolute error (MAE) reduced from 0.1373 V to 53.4 mV.
Highlights the model's potential for integration into battery management systems under dynamic conditions.

Ferahtia et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

February 8, 2026Energy StorageOpen Access

View Full Paper

Ask AI

Bookmark

Authors

Hybrid model improves lithium-ion battery voltage prediction accuracy, highlighting implications for energy storage systems.

This research aims to enhance the accuracy of voltage predictions for lithium-ion batteries using a hybrid modeling approach.
Integrates an equivalent circuit model (ECM) with a multilayer perceptron (MLP) neural network.
Utilizes the Red-Tailed Hawk optimization algorithm to determine ECM parameters.
Evaluates the model using experimental data from a commercial lithium-ion battery under dynamic current profiles.
Conducts an ablation study to assess the impact of network depth on accuracy.
Compares performance against decision tree and random forest algorithms.
Achieves a reduction in prediction error, with RMSE decreasing from 0.1521 V to 66.6 mV.
Mean absolute error (MAE) reduced from 0.1373 V to 53.4 mV.
Highlights the model's potential for integration into battery management systems under dynamic conditions.

Ferahtia et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark