Research

Paper

TESTING March 25, 2026

Conformalized Transfer Learning for Li-ion Battery State of Health Forecasting under Manufacturing and Usage Variability

Authors

Samuel Filgueira da Silva, Mehmet Fatih Ozkan, Faissal El Idrissi, Marcello Canova

Abstract

Accurate forecasting of state-of-health (SOH) is essential for ensuring safe and reliable operation of lithium-ion cells. However, existing models calibrated on laboratory tests at specific conditions often fail to generalize to new cells that differ due to small manufacturing variations or operate under different conditions. To address this challenge, an uncertainty-aware transfer learning framework is proposed, combining a Long Short-Term Memory (LSTM) model with domain adaptation via Maximum Mean Discrepancy (MMD) and uncertainty quantification through Conformal Prediction (CP). The LSTM model is trained on a virtual battery dataset designed to capture real-world variability in electrode manufacturing and operating conditions. MMD aligns latent feature distributions between simulated and target domains to mitigate domain shift, while CP provides calibrated, distribution-free prediction intervals. This framework improves both the generalization and trustworthiness of SOH forecasts across heterogeneous cells.

Metadata

arXiv ID: 2603.24475
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-25
Fetched: 2026-03-26 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.24475v1</id>\n    <title>Conformalized Transfer Learning for Li-ion Battery State of Health Forecasting under Manufacturing and Usage Variability</title>\n    <updated>2026-03-25T16:16:03Z</updated>\n    <link href='https://arxiv.org/abs/2603.24475v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.24475v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Accurate forecasting of state-of-health (SOH) is essential for ensuring safe and reliable operation of lithium-ion cells. However, existing models calibrated on laboratory tests at specific conditions often fail to generalize to new cells that differ due to small manufacturing variations or operate under different conditions. To address this challenge, an uncertainty-aware transfer learning framework is proposed, combining a Long Short-Term Memory (LSTM) model with domain adaptation via Maximum Mean Discrepancy (MMD) and uncertainty quantification through Conformal Prediction (CP). The LSTM model is trained on a virtual battery dataset designed to capture real-world variability in electrode manufacturing and operating conditions. MMD aligns latent feature distributions between simulated and target domains to mitigate domain shift, while CP provides calibrated, distribution-free prediction intervals. This framework improves both the generalization and trustworthiness of SOH forecasts across heterogeneous cells.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.SY'/>\n    <published>2026-03-25T16:16:03Z</published>\n    <arxiv:comment>Submitted to the 2026 American Control Conference (ACC)</arxiv:comment>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Samuel Filgueira da Silva</name>\n    </author>\n    <author>\n      <name>Mehmet Fatih Ozkan</name>\n    </author>\n    <author>\n      <name>Faissal El Idrissi</name>\n    </author>\n    <author>\n      <name>Marcello Canova</name>\n    </author>\n  </entry>"
}