Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HybridoNet-Adapt: A Domain-Adapted Framework for Accurate Lithium-Ion Battery RUL Prediction

Khoa Tran, Bao Huynh, Tri Le, Lam Pham, Vy-Rin Nguyen, Hung-Cuong Trinh, Duong Tran Anh

TL;DR

The paper tackles domain shift in lithium-ion battery remaining useful life prediction by introducing HybridoNet-Adapt, a domain-adaptive framework that learns domain-invariant features via Maximum Mean Discrepancy. It combines a robust feature extractor (LSTM, Multihead Attention, and Neural ODE) with two trainable predictors and a DA loss that balances source and target contributions, enabling accurate RUL predictions without relying on early-cycle data. Extensive experiments on the TRI and LHP datasets show HybridoNet-Adapt consistently outperforms XGBoost, Elastic Net, and state-of-the-art deep learning baselines, with substantial improvements in RMSE, R², and MAPE, and clear evidence that MMD-based DA yields the best transfer performance. The approach offers scalable, real-time potential for battery health management across diverse operating conditions, with future work including self-supervised learning, real-time deployment optimization, and multi-modal data integration.

Abstract

Accurate prediction of the Remaining Useful Life (RUL) in Lithium ion battery (LIB) health management systems is essential for ensuring operational reliability and safety. However, many existing methods assume that training and testing data follow the same distribution, limiting their ability to generalize to unseen target domains. To address this, we propose a novel RUL prediction framework that incorporates a domain adaptation (DA) technique. Our framework integrates a signal preprocessing pipeline including noise reduction, feature extraction, and normalization with a robust deep learning model called HybridoNet Adapt. The model features a combination of LSTM, Multihead Attention, and Neural ODE layers for feature extraction, followed by two predictor modules with trainable trade-off parameters. To improve generalization, we adopt a DA strategy inspired by Domain Adversarial Neural Networks (DANN), replacing adversarial loss with Maximum Mean Discrepancy (MMD) to learn domain-invariant features. Experimental results show that HybridoNet Adapt significantly outperforms traditional models such as XGBoost and Elastic Net, as well as deep learning baselines like Dual input DNN, demonstrating its potential for scalable and reliable battery health management (BHM).

HybridoNet-Adapt: A Domain-Adapted Framework for Accurate Lithium-Ion Battery RUL Prediction

TL;DR

The paper tackles domain shift in lithium-ion battery remaining useful life prediction by introducing HybridoNet-Adapt, a domain-adaptive framework that learns domain-invariant features via Maximum Mean Discrepancy. It combines a robust feature extractor (LSTM, Multihead Attention, and Neural ODE) with two trainable predictors and a DA loss that balances source and target contributions, enabling accurate RUL predictions without relying on early-cycle data. Extensive experiments on the TRI and LHP datasets show HybridoNet-Adapt consistently outperforms XGBoost, Elastic Net, and state-of-the-art deep learning baselines, with substantial improvements in RMSE, R², and MAPE, and clear evidence that MMD-based DA yields the best transfer performance. The approach offers scalable, real-time potential for battery health management across diverse operating conditions, with future work including self-supervised learning, real-time deployment optimization, and multi-modal data integration.

Abstract

Accurate prediction of the Remaining Useful Life (RUL) in Lithium ion battery (LIB) health management systems is essential for ensuring operational reliability and safety. However, many existing methods assume that training and testing data follow the same distribution, limiting their ability to generalize to unseen target domains. To address this, we propose a novel RUL prediction framework that incorporates a domain adaptation (DA) technique. Our framework integrates a signal preprocessing pipeline including noise reduction, feature extraction, and normalization with a robust deep learning model called HybridoNet Adapt. The model features a combination of LSTM, Multihead Attention, and Neural ODE layers for feature extraction, followed by two predictor modules with trainable trade-off parameters. To improve generalization, we adopt a DA strategy inspired by Domain Adversarial Neural Networks (DANN), replacing adversarial loss with Maximum Mean Discrepancy (MMD) to learn domain-invariant features. Experimental results show that HybridoNet Adapt significantly outperforms traditional models such as XGBoost and Elastic Net, as well as deep learning baselines like Dual input DNN, demonstrating its potential for scalable and reliable battery health management (BHM).

Paper Structure

This paper contains 22 sections, 18 equations, 11 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Motivations
  3. Problem Statement
  4. Main Contribution
  5. Preliminaries
  6. LSTM graves2012long:
  7. Multihead Attention vaswani2017attention:
  8. NODE chen2018neural:
  9. Proposed method
  10. Overall architecture
  11. HybridoNet-Adapt: A Proposed RUL Prediction model with Novel Domain Adaptation
  12. Experiments and discussion
  13. Experimental Setup
  14. Datasets
  15. First dataset: Varied fast-charging conditions, with consistent discharging conditions
  16. ...and 7 more sections

Figures (11)

  • Figure 1: The overall RUL prediction process for Lithium-ion battery cells.
  • Figure 2: Architecture of the proposed HybridoNet-Adapt framework for RUL prediction with domain adaptation.
  • Figure 3: Comparison of maximum charge and discharge capacities over cycle life for battery cells.
  • Figure 4: Comparison of RUL prediction performance with and without median filtering.
  • Figure 5: Performance comparison of LSTM-based blocks for feature extractor. Experiment on the testing data of the second dataset.
  • ...and 6 more figures