Early Remaining Useful Life Prediction of Lithium-Ion Batteries Based on a Hybrid Machine Learning Method with Time Series Augmentation

TL;DR

This paper introduces a hybrid machine learning method to accurately predict lithium-ion battery life early on, even with limited and noisy data.

Contribution

A novel hybrid framework combining signal decomposition, time series augmentation, and deep forecasting for early battery RUL prediction.

Findings

The proposed method achieves high accuracy with R2 scores between 0.9643 and 0.9972 using only 20% of historical data.

RMSE and MAE values are below 0.0296 and 0.0198, respectively, indicating strong predictive performance.

The framework is robust to noise and data scarcity in early-stage battery life prediction.

Abstract

Early and accurate prediction of the remaining useful life (RUL), defined as the number of operational cycles a battery can continue to function before reaching its end-of-life threshold, is crucial for improving the reliability of new energy vehicles. To address noise contamination, capacity regeneration effects, and data scarcity in early-stage prognostics, this paper proposes a hybrid framework integrating signal decomposition, time series augmentation, and deep forecasting. The raw capacity sequence is decomposed using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to separate multi-scale components. A Transformer-enhanced time series generative adversarial network (HyT-GAN) is then employed to augment decomposed components, improving robustness under small-sample conditions. A CNN-BiGRU predictor is trained for capacity forecasting, and key hyperparameters are tuned via the Dung Beetle Optimizer (DBO). Experiments on NASA and CALCE benchmark datasets demonstrate that the proposed method achieves accurate early-stage prediction using only 20% historical data, with R2 ranging from 0.9643 to 0.9972 and RMSE/MAE below 0.0296/0.0198. These results indicate that the proposed framework can deliver reliable RUL estimates under data-limited and noisy measurement conditions.

Figures