NOCE-Net learns battery health from raw operational data
NOCE-Net: Representation Learning for Battery Operational Context via Nested Sequence Modelling
Computational Engineering, Finance, and Science
Summary
Battery health depends on how it is used over time, but existing methods often need specific features or early data, making it hard to predict future battery performance. The authors created NOCE-Net, a deep learning model that understands battery operation directly from raw data without handcrafted features. They show it works well on many battery types and usage patterns, even ones it never saw before, with high accuracy and fewer model parameters. The model also captures battery degradation trends, which can help predict battery lifespan and performance in real-world applications.
What this means in practice
- •For battery management engineers: Predict battery voltage responses and degradation trajectories using raw operational data across diverse chemistries without requiring future response data.
- •For electric vehicle maintenance teams: Monitor battery health in vehicles under varying usage patterns, improving maintenance scheduling with generalized predictions from raw telemetry.
Authors
Arnab Bhattacharjee, Wayes Tushar, Tapan Kumar Saha
Abstract
Learning effective representations from battery operational history is essential for predictive health management and control-oriented applications. Existing approaches rely on response-dependent engineered features or short early-life windows, limiting generalisability and precluding use in prospective settings where future responses are unavailable. We propose NOCE-Net, a deep nested sequence model that learns control-enabling representations directly from raw cyclic operational or telemetric data. Our data pipeline, re-BatteryData, standardises heterogeneous aging datasets into semantically equivalent Full Equivalent Cycles (FECs) defined by absolute charge throughput, enabling end-to-end learning across diverse operational regimes. NOCE-Net combines multi-patch transformer encoders for intra-cycle dynamics with a GRU backbone for inter-cycle state evolution, operating on approximately $700$ batteries across ten public datasets spanning over $300$ operational profiles. In zero-shot evaluation on unseen battery chemistries and usage patterns, NOCE-Net achieves $13.7\%$ MAPE in voltage response prediction, outperforming strong long-sequence baselines by over $43\%$ while using $40 \times$ fewer parameters. Shape and component diagnostics of the learned hidden states reveal that the GRU's principal direction of evolution achieves Spearman $|ρ| \geq 0.95$ with the capacity fade trajectory on average across all datasets with monotone degradation profiles, including zero-shot test batteries, providing evidence that the representations encode degradation dynamics transferable to downstream tasks beyond battery response modelling. Code is available at https://github.com/BeArnab96/NOCE-Net.