2026· Journal of Energy Engineering· Vol 152· 0 citations· 15 references
TL;DR
A reinforcement learning driven deep temporal network (RL-DTNet) is proposed for SoC prediction, integrating of reinforcement learning for self-correction, temporal attention to handle dynamic dependencies, and a degradation-aware model for long-term prediction accuracy.
Abstract
Accurate state-of-charge (SoC) estimation is essential for ensuring the reliability and operation of lithium-ion batteries in electric vehicles. Traditional data-driven models failed to generalize across varying load profiles. Also, they do not handle nonlinear electrochemical behavior and long-term degradation effects of batteries. Recently, reinforcement learning–based SoC correction techniques have shown promising results in SoC predictions. But these models are based on synthetic or simplified feedback, which is unstable for real-world conditions. The goal of this work is to develop a novel approach for improving SoC prediction in real-time battery management systems. To achieve this, in this work, a reinforcement learning driven deep temporal network (RL-DTNet) is proposed for SoC prediction. The RL-DTNet consists of four major stages: (1) an adaptive feature extraction layer to construct an enriched multidomain representation; (2) a temporal attention gated recurrent unit (GRU) module to focus on the most informative timesteps under varying load disturbances; (3) a reinforcement learning self-correction module using a deep Q-network to learn how to minimize prediction errors by adjusting outputs based on real-time error feedback; and (4) a long-term degradation modeling layer to adjust SoC estimates for aging effects using a cycle-aware degradation factor. The novel contribution of this work is the integration of reinforcement learning for self-correction, temporal attention to handle dynamic dependencies, and a degradation-aware model for long-term prediction accuracy. Experimental results on real-time datasets show that RL-DTNet achieves better results of the root mean squared error, mean absolute error, and coefficient of determination (
R
2
) of 0.6377, 0.5056, and
R
2
, respectively.
Accurate state of charge (SoC) estimation is of primary importance in terms of safe and efficient management of energy storage systems (ESSs). In this regard, data-driven frameworks offer the advantage of rapid execution during online operations. Nevertheless, their deployment on resource-constrained embedded systems is often hindered by the strict memory and processing limitations of low-cost hardware. This article proposes a three-stage pipelined SoC estimation framework incorporating a reinforcement learning (RL) primary stage, least squares boosting (LSB) secondary residual corrector, and ultimate linear three-point interpolation (3p-InT) stage. The RL phase utilizes a twin deep delayed deterministic policy gradient neural network (TD3NN)-based agent along with a customized reward function. The inference part of the framework is deployed on two separate embedded platforms, i.e., an STM32F411RE microcontroller (MCU) and Digilent Nexys A7-100T FPGA through automatic C code and hardware description language (HDL) code generation features in MATLAB/Simulink, respectively. The efficacy of the proposed framework is evaluated using a Panasonic 18650 lithium-ion battery (LiB) and a battery-powered drill load profile (BPD-LP). Across the four hardware scenarios, the accuracy of the proposed framework is preserved, with the maximum %RMSE deviation not exceeding 0.08 percentage points. The RMSE value remains within 1.80–1.82% for the LiB dataset and within 0.76–0.84% for the BPD-LP, irrespective of the platform or the arithmetic format. As for the resource footprint, the fixed-point implementation more than halves the FPGA logic with respect to the floating point (27.33% against 65.55% of the LUTs), at the cost of a comparatively higher DSP usage (15% against 7.08%). On the MCU, it trades additional flash memory (31.51% against 25.23%) for a 2.6-fold smaller RAM footprint. The framework’s reward function, hyperparameters, and architecture are kept unchanged across both datasets, indicating that the same configuration can be generalized across both profiles without requiring dataset-specific re-tuning. Moreover, the detailed hardware deployment findings provide a practical insight into the hardware and arithmetic format selection for an accurate embedded SoC estimation framework.
Accurate prediction of the remaining useful life (RUL) of lithium-ion batteries (LIBs) is essential for enhancing the reliability, operational safety, and energy management efficiency of electric vehicles and modern energy storage systems. However, battery degradation is governed by highly nonlinear electrochemical mechanisms and complex temporal dependencies that are difficult to model using conventional physics-based approaches or standalone machine learning techniques. To address these challenges, this article proposes a hybrid data-driven framework integrating a Temporal Convolutional Network (TCN), Bidirectional Long Short-Term Memory (BiLSTM), and Extreme Gradient Boosting (XGBoost) for accurate LIB RUL prediction. The proposed architecture utilizes the TCN module to capture short-term temporal degradation patterns from sequential battery operational data, while the BiLSTM network learns long-term temporal dependencies and degradation evolution across multiple charge–discharge cycles. The deep temporal representations extracted by the TCN–BiLSTM network are subsequently processed using an XGBoost regression model to effectively model nonlinear relationships between battery operational characteristics and RUL. The framework is validated using the NASA LIB aging dataset containing 14,896 charge–discharge cycle samples with operational features including cycle index, discharge time, charging duration, voltage degradation characteristics, and constant-current charging behavior. Statistical analysis and Min–Max normalization are employed to improve feature consistency, numerical stability, and model convergence. Experimental results demonstrate that the proposed framework effectively captures battery degradation dynamics and achieves highly accurate and stable prediction performance. Five-fold cross-validation results yield a low Mean Absolute Error of 0.00957, Root Mean Square Error of 0.02926, and a high coefficient of determination (
R
2
) of 0.9882, indicating excellent predictive capability and strong generalization performance. Comparative analysis further demonstrates that the proposed hybrid framework outperforms conventional Random Forest, XGBoost, LSTM, and BiLSTM models in terms of prediction accuracy and robustness. In addition, ablation analysis confirms the complementary contribution of temporal convolutional learning, sequential dependency modeling, and ensemble nonlinear regression toward improved RUL estimation. The proposed framework provides a robust and computationally efficient solution for intelligent battery health monitoring, predictive maintenance, and smart battery management applications in electric vehicles and energy storage systems.
T. Mariprasath, Kumaresh S S, Seif Al Bustanji et al.· Energy Exploration & Exp...· 0 citations
Accurate State of Charge (SoC) estimation is essential for safe and efficient lithium-ion battery operation in electric mobility. Although machine learning methods achieve high predictive capability, many studies rely on laboratory cycling data and overlook deployment constraints such as latency and embedded hardware. This paper presents a deployment-oriented evaluation of data-driven SoC estimation using large-scale operational data from a battery electric bus. The dataset includes more than 4.3 million field measurements of current, voltage, temperature, and SoC collected under realistic driving conditions. Five models are compared: Random Forest, LightGBM, XGBoost, Temporal Convolutional Networks (TCN), and Gated Recurrent Unit (GRU) networks. Models are evaluated across sampling intervals from 30 s to 210 s and multiple training-data fractions, considering accuracy, training time, and inference latency. Results show that temporal resolution strongly affects the accuracy-efficiency trade-off. The best configuration, LightGBM at 180 s using 75 % of training data, achieved a test MAE of 6.338, RMSE of 8.495, and median inference latency of 1.10 ms. Compared with the Random Forest baseline at 30 s, it reduced MAE by 34.4 %, RMSE by 35.6 %, and latency by 97.6 %, supporting efficient real-time battery management system deployment.
J. Orellana-Iñiguez, W. L. Gallo, Madon C. de Almeida· Eletrônica de Potência· 0 citations
With the increasing integration of renewable energy sources, energy storage systems have become essential, making the accurate estimation of their State of Health (SOH) and degradation behavior critical. In this work, we propose a physics-informed deep learning approach for lithium-ion battery SOH prediction using incomplete discharge curves extracted from arbitrary voltage ranges, thereby reflecting realistic and heterogeneous operating conditions. The proposed method combines data-driven learning with physically motivated degradation dynamics to ensure consistent and reliable SOH estimation from partial discharge information, achieving a MAPE below 4$\%$. In addition, a real-time degradation trend estimation strategy is introduced to detect key aging transitions without requiring prior knowledge or historical data, making it applicable to a wide range of batteries. Overall, our approach enables SOH estimation from arbitrary discharge segments and a real-time degradation forecast that continuously integrates all usage, overcoming previous methods that rely on fixed protocols or early, non-adaptive predictions.
Begoña Ispizua, S. Gil-Lopez, Leire Arrizabalaga et al.· 0 citations
A fresh enhanced training set covering all-round driving cycles based on real driving data is constructed, thus breaking through the training restrictions brought by standard driving cycles, and improves fuel economy by 5.2% on average across various driving cycles compared to adaptive ECMS (AECMS).
Yushan Li, Lianbo Zhao, Fanyu Meng et al.· Proceedings of the Instituti...· 0 citations
The efficient operation of lithium-ion battery management systems (BMSs) depends on accurate state-of-charge (SOC) estimation. However, the performance of conventional model-based SOC estimation methods may progressively worsen owing to parameter uncertainty and nonlinear battery dynamics. This study proposes a hybrid SOC estimation framework termed DO-EKFRes, comprising two sequential stages. In the first stage, the process and measurement-noise covariance matrices are optimized offline using a data-driven strategy. In the second stage, a Bidirectional Long Short-Term Memory (BiLSTM) residual learning network is employed to compensate for the remaining SOC estimation errors. The proposed framework was evaluated using two complementary validation protocols: a synthetic Monte Carlo experiment and a Leave-One-Battery-Out (LOBO) cross-validation framework based on the NASA Prognostics Center of Excellence (PCoE) lithium-ion battery dataset. In the synthetic validation, DO-EKFRes achieved an RMSE of 0.803%, corresponding to reductions of 48.83% and 26.84% relative to the EKF and DO-EKF, respectively. In the NASA LOBO evaluation, the proposed framework achieved a macro-averaged RMSE of 11.534%, corresponding to reductions of 49.25% and 9.68% relative to the EKF and DO-EKF, respectively. These results demonstrate that integrating offline covariance optimization with BiLSTM-based residual learning improves estimation accuracy, robustness, and cross-battery generalization, providing a practical solution for lithium-ion battery SOC estimation in battery management systems.
M. I. al-Hiyali, Y. Naif, Ramani Kannan et al.· World Electric Vehicle Journ...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
