BatteryMFormer: Multi-level Learning for Battery Degradation Trajectory Forecasting

2026-05-26

AI Summary

Accurately predicting the full-lifecycle degradation trajectory of battery state of health at an early stage remains challenging, as existing methods struggle to effectively model multi-level data structures and capture local dynamic characteristics within specific state-of-charge (SOC) intervals. To address this, this work proposes BatteryMFormer—a novel multi-level Transformer architecture that integrates aging-condition priors, a cross-battery meta-degradation memory mechanism, and an SOC-local-sensitive dual-view temporal encoder. The model jointly captures both shared degradation patterns and individual battery variations through an aging-condition-aware decoder, a meta-memory module, and a dual-view encoder. Extensive experiments on four public battery datasets demonstrate that BatteryMFormer significantly outperforms state-of-the-art methods, confirming its effectiveness in enabling early, high-accuracy, and generalizable degradation trajectory prediction.

Abstract

Early battery degradation trajectory forecasting (BDTF), which predicts the full-life state-of-health trajectory from early operational data, is critical for battery optimization, manufacturing, and deployment. Battery degradation data exhibit two key characteristics. First, degradation data present a multi-level structure, including regularities shared within aging conditions and trajectory patterns shared across batteries. Second, degradation-related variations in voltage-current profiles are often localized to specific state-of-charge (SOC) intervals. Existing approaches often fail to explicitly model these characteristics. To bridge this gap, we propose BatteryMFormer, a multi-level Transformer for early BDTF. BatteryMFormer integrates (1) an aging-condition-aware decoder that injects aging-condition priors via aging-condition-informed queries and aging-condition-aware attention, (2) a meta degradation pattern memory that learns and retrieves trajectory prototypes to guide long-horizon forecasting, and (3) a dual-view encoder that jointly captures temporal dynamics and SOC-localized variations from voltage and current time series. Extensive experiments on four battery domains show that BatteryMFormer consistently outperforms state-of-the-art baselines, marking a significant step toward reliable BDTF. Our code is available at https://github.com/Ruifeng-Tan/BatteryMFormer.

Problem

• battery degradation trajectory forecasting
• multi-level structure
• state-of-charge (SOC) intervals
• early prediction
• voltage-current profiles

Innovation

• multi-level Transformer
• battery degradation trajectory forecasting
• aging-condition-aware attention
• meta degradation pattern memory
• SOC-localized variations