cs.LGOct 25, 2026

Single-Cycle Battery Remaining Useful Life Prediction with Compressed Continuous Wavelet Features and Deep Neural Networks

Authors: Sungeun Kim, Mingyu Jung, Yuri Kim, Jin-Sung Park, Jongbin Ryu

Organizations: Department of Artificial Intelligence, Ajou University, Suwon, 16499, Republic of Korea · Department of Materials Science & Engineering, Ajou University, Suwon, 16499, Republic of Korea · Department of Software & Computer Engineering, Ajou University, Suwon, 16499, Republic of Korea

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Lithium-ion batteries are widely used in industrial applications, but their performance gradually degrades over time, making accurate lifetime prediction important for safety and reliability. However, reliable lifetime assessment generally requires extensive degradation testing, which is both costly and time-consuming. To reduce this burden, recent data-driven approaches have sought to predict the end-of-life cycle of batteries from only the initial few cycles. Nevertheless, identifying reliable degradation patterns from such limited early-cycle data remains challenging because early aging dynamics are often subtle and nonlinear. To address this issue, we adopt the continuous wavelet transform to convert cycle-wise voltage, current, and capacity signals into time-frequency representations, enabling multi-resolution characterization of degradation behavior. Building on this representation, we apply scale compression to reduce redundancy while preserving structural information in the scalograms. We evaluate the proposed framework on eight benchmark datasets derived from seven public battery datasets covering diverse electrode chemistries, cell form factors, and cycling protocols. Experimental results show that the proposed framework consistently outperforms prior approaches in single-cycle remaining useful life prediction and remains robust as the number of input cycles increases from 1 to 100. In addition, redundancy-fidelity analysis characterizes the trade-off introduced by scale compression, and gradient-based visual explanations indicate that the model focuses on localized time-frequency regions associated with structured signal transitions during charge-discharge operation. These results demonstrate the effectiveness and interpretability of the proposed framework for early-cycle battery remaining useful life prediction.

Figures & tables

Figure 1: Framework for RUL prediction of lithium-ion batteries. Charge/discharge time-series (voltage, current, capacity) are converted to time-frequency scalograms using the continuous wavelet transform, followed by scale compression to reduce redundancy. The compressed features are fed to a deep neural network for RUL prediction.
CharacteristicsCALCEHNEIMATRUL_PURSNLRWTHHUST
Anode materialgraphitegraphitegraphitegraphitegraphitegraphitegraphite
Cathode materialLCONMC/LCOLFPNCANCA/NMC/LFPNMCLFP
Form factorPrismaticCylindricalCylindricalCylindricalCylindricalCylindricalCylindrical
Voltage range (V)2.7-4.23.0-4.32.0-3.52.7-4.22.0-4.23.5-3.92.0-3.6
Nominal capacity (Ah)1.230.121.23_{0.12}2.8002.80_{0}1.1001.10_{0}3.4003.40_{0}2.390.882.39_{0.88}1.1101.11_{0}1.101.1_{0}
RUL distribution (cycles)566106566_{106}24815248_{15}823368823_{368}20950209_{50}125613211256_{1321}65864658_{64}18993891899_{389}
Table 1: Characteristics of the battery datasets used for RUL prediction. The subscripts denote the standard deviation.
Figure 2: Degradation trajectories and EOL distributions of cells in (a) MATR1 and (b) MIX benchmarks.
ModelCycleMATR1MATR2HUSTSNLCLOCRUHCRUSHMIXAvg
Ridge regression2356.27500.72>> 1000417.92515.33325.29601.98576.19576.24
PCR2383.80514.74419.91458.45318.31182.59569.58567.52426.86
PLSR2374.91521.74431.05476.05313.76173.32567.56581.83430.03
Gaussian process2356.27500.73>> 1000406.84515.33199.14>> 1000>> 1000>> 1000
XGBoost2330.50352.60587.05489.28291.38152.43531.99381.00389.53
Random forest2327.72363.39447.37401.77286.46159.75531.95407.33365.72
Table 2: Remaining useful life prediction results on benchmark datasets, evaluated at each method’s minimum number of input cycles. “ >> 1000” indicates that the RMSE exceeds 1000, reflecting large prediction errors. The subscripts denote the standard deviation.
modelCycleRMSEMAEMAPE
Ridge regression2576.24371.400.51
PCR2426.86306.490.54
PLSR2430.03298.710.48
Gaussian process2>1000>1000>1.00
XGBoost2389.53254.240.38
Random forest2365.72232.900.29
Table 3: Comparison of average RMSE, MAE, and MAPE across all benchmark datasets, using the fewest input cycles each method requires.
Figure 3: Heatmap showing average RMSE, MAE, and MAPE across benchmarks as the number of input cycles increases. White cells indicate missing settings (e.g., methods requiring ≥2\geq 2 or ≥5\geq 5 cycles). Our method demonstrates robustness across varying numbers of input cycles.
Figure 4: Prediction results on the MIX dataset for BatLiNet and our method at their minimum input (top row; 2 and 1 cycles, respectively) and at 100 cycles (bottom row).
Figure 5: Grad-CAM visualization results for different input channels. (a) Voltage, (b) Current, and (c) Capacity. Each panel shows the original signal, the corresponding CWT scalogram, and the Grad-CAM attention map.
Input signalsRMSEMAEMAPE
V184.07126.880.15
I182.30124.930.14
C199.36138.980.19
IC186.10129.610.15
VC191.34145.800.19
VI186.36130.240.15
Table 4: RUL prediction performance on the MIX dataset under different input signal configurations. V, I, and C denote voltage, current, and capacity, respectively.
MethodRMSEMAEMAPE
Time-domain signal274.3192.50.27
STFT251.4174.20.29
DWT254.2177.60.25
CWT227.7152.00.23
Table 5: Comparison of time-frequency feature extraction methods for battery RUL prediction under the 1-cycle setting.
Figure 6: Comparison of average prediction error across three input configurations: time-domain, CWT, and CWT with scale compression.
CompressionCycleParameters (M)GFLOPsTraining time (s)
 111.470.40313.48
✓111.320.36225.84
 10013.7338.431522.05
✓10011.324.56240.06
Table 6: Comparison of model complexity and training cost with and without scale compression under the 1-cycle and 100-cycle settings. Parameters (M) denotes the number of trainable model parameters in millions, and GFLOPs are reported on a per-sample basis.
Figure 7: Auxiliary analysis of knee-point prediction under the single-cycle setting.
Figure 8: Trade-off between redundancy ratio and fidelity under different scale compression levels.
MethodCycleRMSEMAEMAPE
PCA1331.33233.770.43
100266.65183.250.33
Top- kk energy1309.75218.770.48
100294.99213.150.38
Energy-weighted1249.28167.140.24
100233.43159.370.32
Table 7: Ablation results for different scale compression methods.
Figure 9: Trade-off between redundancy and fidelity across different scale compression strategies.
More in this topic

RUL Estimation

Remaining useful life estimation predicts how long machines, batteries, and other assets can operate before failure or end of life.

See all 57 papers

Explore similar work

CardsList
Apr 17, 2026

Continuous ageing trajectory representations for knee-aware lifetime prediction of lithium-ion batteries across heterogeneous dataset

Accurate assessment of lithium-ion battery ageing is challenged by cell-to-cell variability, heterogeneous cycling protocols, and limited transferability of data-driven models across datasets. In particular, robust identification of degradation transitions, such as the knee point, and reliable early-life prediction of remaining useful life (RUL) remain open problems. This study proposes a unified framework for battery ageing analysis based on continuous representations of voltage-capacity and capacity-cycle trajectories learned from heterogeneous public datasets (NASA, CALCE, ISU-ILCC). The continuous formulation enables consistent extraction of degradation descriptors, including curvature, plateau length and knee-related metrics, while reducing sensitivity to dataset-specific discretisation. Across more than 250 cells, statistically significant correlations between knee onset and end-of-life (Pearson 0.75-0.84) are observed. Additional early-life analysis confirms that knee-related features retain predictive value when estimated from partial trajectories. Early-life models provide increasingly stable RUL predictions as the number of observed cycles increases, with meaningful predictive performance emerging within the first 5-20 cycles and remain robust under cross-dataset domain shift. The framework integrates continuous modelling, feature extraction and uncertainty-aware prediction, providing an interpretable and dataset-consistent approach demonstrating robustness across heterogeneous dataset types. Compared with conventional discrete or feature-based methods, the proposed representation reduces sensitivity to sampling resolution and improves cross-dataset consistency. The study is limited to laboratory-scale datasets and capacity-based end-of-life definitions.

Agnieszka Pregowska, Stefan Marynowicz

RUL EstimationDomain GeneralizationTime Series Representation LearningLithium-Ion Batteries
AskPDF
Sep 18, 2026

Joint Remaining Useful Life Prediction and Capacity Estimation of Lithium-Ion Batteries Using Partial-Charging Data

Joint remaining useful life (RUL) prediction and capacity estimation require representations of both gradual degradation and recent battery behavior. This paper presents a cross-expert framework using partial-charging measurements without requiring measured historical full-cycle capacity as an input. The RUL Expert captures long-term degradation from nominal 10-min segments sampled across a 30-cycle history, while the Capacity Expert characterizes recent battery behavior from statistical descriptors of nominal 40-min segments over ten consecutive cycles. Their complementary representations are integrated through feature-wise linear modulation for joint RUL and capacity prediction. A key contribution is a three-stage training strategy that progressively controls frozen and trainable components: supervised representation pretraining, independent expert pretraining, and final fusion training with both experts frozen. This staged optimization preserves expert-specific degradation knowledge while improving the balance between the two prediction tasks, with RUL treated as the primary prognostic objective. On two public battery-aging datasets, the reference configuration achieves mean RUL root-mean-square errors of 143.69 and 161.10 cycles and capacity errors of 12.36 and 7.28 mAh, respectively. On Dataset I, cross-expert fusion reduces both mean errors relative to either standalone expert. The proposed framework achieves the lowest reported RUL RMSE among the compared methods on both datasets while maintaining competitive capacity-estimation accuracy.

Khoa Tran, Ho-Si-Hung Nguyen, Phone Wai Yan Moe +2

Data Science Laboratory, Faculty of Information Technology, Ton Duc Thang University, Ho Chi Minh City 70000, Vietnam · Faculty of Electrical Engineering, The University of Danang - University of Science and Technology (DUT), 54 Nguyen Luong Bang Street, Lien Chieu District, Da Nang 550000, Vietnam · AIWARE Limited Company, 17 Huynh Man Dat Street, Hoa Cuong Ward, Hai Chau District, Da Nang 550000, Vietnam +2

RUL EstimationMulti-Task LearningLithium-Ion Batteries
AskPDF
Mar 28, 2026

Hybrid Deep Learning with Temporal Data Augmentation for Accurate Remaining Useful Life Prediction of Lithium-Ion Batteries

Accurate prediction of lithium-ion battery remaining useful life (RUL) is essential for reliable health monitoring and data-driven analysis of battery degradation. However, the robustness and generalization capabilities of existing RUL prediction models are significantly challenged by complex operating conditions and limited data availability. To address these limitations, this study proposes a hybrid deep learning model, CDFormer, which integrates convolutional neural networks, deep residual shrinkage networks, and Transformer encoders extract multiscale temporal features from battery measurement signals, including voltage, current, and capacity. This architecture enables the joint modeling of local and global degradation dynamics, effectively improving the accuracy of RUL prediction.To enhance predictive reliability, a composite temporal data augmentation strategy is proposed, incorporating Gaussian noise, time warping, and time resampling, explicitly accounting for measurement noise and variability. CDFormer is evaluated on two real-world datasets, with experimental results demonstrating its consistent superiority over conventional recurrent neural network-based and Transformer-based baselines across key metrics. By improving the reliability and predictive performance of RUL prediction from measurement data, CDFormer provides accurate and reliable forecasts, supporting effective battery health monitoring and data-driven maintenance strategies.

Yun Tian, Guili Wang, Jian Bi +7

School of Physics and Electronic Information, Anhui Normal University, Wuhu 241002, China · College of Mechanical and Vehicle Engineering, Hunan University, Changsha 410082, China · School of Design, NingboTech University, Ningbo 315100, China +2

RUL EstimationLithium-Ion BatteriesHybrid CNN-Transformer ArchitecturesTime Series Data Augmentation
AskPDF