SambaMixer: State of Health Prediction of Li-ion Batteries using Mamba State Space Models

TL;DR

This work tackles Li-ion battery SOH prediction from multi-variate time-series by introducing SambaMixer, a Mamba-based structured state-space approach that delivers channel-aware processing with sub-quadratic complexity. It couples anchor-based resampling with sample-time and cycle-difference positional embeddings to handle varying sequence lengths and recuperation effects, enabling accurate cycle-level SOH estimation ($SOH_k$). Empirical results on the NASA PCoE discharge dataset show SambaMixer achieving superior MAE, RMSE, and MAPE compared to the prior state-of-the-art, and robust performance across used-battery scenarios and data splits. The approach supports long-range dependency modeling suitable for battery management systems, though validation is limited to a single dataset and future work should extend to diverse chemistries and discharge profiles.

Abstract

The state of health (SOH) of a Li-ion battery is a critical parameter that determines the remaining capacity and the remaining lifetime of the battery. In this paper, we propose SambaMixer a novel structured state space model (SSM) for predicting the state of health of Li-ion batteries. The proposed SSM is based on the MambaMixer architecture, which is designed to handle multi-variate time signals. We evaluate our model on the NASA battery discharge dataset and show that our model outperforms the state-of-the-art on this dataset. We further introduce a novel anchor-based resampling method which ensures time signals are of the expected length while also serving as augmentation technique. Finally, we condition prediction on the sample time and the cycle time difference using positional encodings to improve the performance of our model and to learn recuperation effects. Our results proof that our model is able to predict the SOH of Li-ion batteries with high accuracy and robustness.

Figures (13)

• Figure 1: Effect of battery aging on the measured voltage, current and temperature of various discharge cycles of a Li-ion battery. Battery #5 of NASA's battery dataset saha2007nasa.
• Figure 2: SambaMixer architecture. We input a multi-variate time series of current, voltage, temperature and sample time. We first first resample the time signals using our anchor-based resampling technique. We then feed the resampled sample time into the sample time positional encoding layer. We further feed the time difference between two discharge cycles in hours into the cycle time difference positional encoding layer. The other signals, i.e. current, voltage and temperature are fed into the input projection. The projected signals are added to the sample time embeddings and the cycle time difference embeddings. Optionally, a CLS token can be inserted at any position. The embedded tokens are then fed into the SambaMixer Encoder. The SambaMixer Encoder consists of $M$ stacked SambaMixer Encoder blocks. The output of the encoder is finally fed into the head, which predicts the state of health of the current cycle $k$ for battery $b_{\psi}$.
• Figure 3: Resample techniques. Original: The original sample time sequence with $L_{k}^{\psi}$ samples. Linear: linear resampling with $L$ equidistant samples. Random: random resampling with $L$ samples drawn from a uniform distribution. Anchor: anchor-based resampling with random uniform noise $z$ added to $L$ equidistant samples.
• Figure 4: Capacity degradation for all selected batteries.
• Figure 5: SOH prediction for Battery #06