Accurate estimating the state-of-charge (SOC) of Li-ion battery contributes significantly to electric vehicle safety. Existing methods typically focus on the traditional recurrent neural networks to encode time series features for SOC estimation. However, these methods rely solely on their own structure to extract time series correlated features, ignoring a significant amount of information on temporal dimension. To address this issue, this paper proposes a temporal transformer-based sequence network (TTSNet) that can make full use of temporal dimensional information to model the relationship between the input and SOC. Specifically, the proposed network splits the raw data into three branches including voltage, current, and temperature, as well as extracts the corresponding primary semantic features. It then uses a temporal transformer to effectively encode the features of temporal dimensional information. The resulting features are further fed into an attention-guided feature fusion module to interact information among voltage, current, and temperature branches for subsequent SOC estimation. To enhance the network's resilience for long time sequences, a sliding time window technique is introduced to pre-process the raw data. Besides, a Kalman filter is incorporated as post-processing to smooth the output to guide a more accurate SOC estimation. Comprehensive experiments are conducted on battery open datasets and vehicle operation datasets to verify the proposed method. The results demonstrate that the proposed method achieves high accuracy and strong robustness in both datasets, with average MAE, RMSE, and $\mathrm{R^{2}}$ values of 0.506%, 0.694%, and 99.791%, respectively. The code is available at https://github.com/haooozi/TTSNet .

Accurate estimation of lithium-ion battery state-of-health (SOH) is important for the safe operation of electric vehicles; however, in practical applications, the accuracy of SOH estimation is affected by uncertainty factors, including human operation, working conditions, etc. To accurately estimate the battery SOH, a hybrid neural network based on the dilated convolutional neural network and the bidirectional gated recurrent unit, namely dilated CNN-BiGRU, is proposed in this paper. The proposed data-driven method uses the voltage distribution and capacity changes in the extracted battery discharge curve to learn the serial data time dependence and correlation. This method can obtain more accurate temporal and spatial features of the original battery data, resulting higher accuracy and robustness. The effectiveness of dilated CNN-BiGRU for SOH estimation is verified on two publicly lithium-ion battery datasets, the NASA Battery Aging Dataset and Oxford Battery Degradation Dataset. The experimental results reveal that the proposed model outperforms the compared data-driven methods, e.g., CNN-series and RNN-series. Furthermore, the mean absolute error (MAE) and root mean square error (RMSE) are limited to within 1.9% and 3.3%, respectively, on the NASA Battery Aging Dataset.

State-of-X (SOX) estimation of lithium-ion batteries is crucial for safe operation of electric vehicles (EVs). However, EVs have long suffered from complex and variable operation conditions. While deep learning-based state estimation demonstrates strong generalization to such operation conditions, it typically focus on estimating a single state, and is evaluated on simulated datasets such as CALCE. In this paper, we introduce a dual-task learning framework for joint state-of-charge (SOC) and state-of-energy (SOE) estimation of lithium-ion battery pack, and verify it on real vehicle data. This novel framework possesses two appealing properties: 1) It incorporates a feature attention mechanism to capture task-relevant temporal features encoded by a gated recurrent unit. 2) It leverages data preprocessing operations, including correlation analysis and sliding windows, enhancing both efficiency and accuracy of the model. Comprehensive experiments are conducted on actual operation data from six EVs with a cumulative mileage exceeding 80,000 kilometers. These data are further categorized into early, mid, and late stages based on the battery’s health status. The experimental results show that our method achieves SOC and SOE errors of less than 3%, verifying the high accuracy and robustness of the proposed framework under complex and variable vehicle operating conditions.

The time-varying, dynamic, nonlinear, and other characteristics of lithium-ion batteries, as well as the capacity regeneration phenomenon, leads to the low accuracy of the traditional deep learning models in predicting the remaining useful life of lithium-ion batteries. This paper established a sequence-to-sequence model for remaining useful life prediction by combining the variational modal decomposition with bi-directional long short-term memory and Bayesian hyperparametric optimization. First, variational modal decomposition is used for noise reduction processing to maximize the retention of the original information of capacity degradation. Second, the capacity declining trend after noise reduction is modeled and predicted by the combination of bi-directional long-short term memory and temporal attention mechanism. In addition, a Bayesian optimizer is used to adaptively adjust the hyperparameters while training the model. Finally, the model was validated on NASA and CALCE data sets, and the results show that the model can accurately predict the future trend with only the initial 12% capacity data.

The precise estimation of the state-of-health (SOH) in lithium-ion battery holds significant importance for ensuring the safe operation of electric vehicles (EVs). While existing methods predominantly center around conventional recurrent neural networks (RNNs), these approaches inherently encounter challenges in effectively extracting features and modeling ultra-long-term time-series data. In this paper, we introduce a novel time-series prediction framework that integrates feature engineering with an attentive graph neural network (GNN) to enhance the accuracy and efficiency of SOH estimation for lithium batteries. Specifically, we begin by extracting health factors (HFs) from the original charge and discharge data of the battery. Subsequently, we identify 3D features exhibiting strong correlations, which serve as the primary inputs for the network. Our proposed network, termed the Feature Engineering-Guided Attentive Graph Neural Network (FGNet), is designed to comprehensively capture and model the intricate relationships between HFs and SOH output. This is achieved by embedding both a feature extraction module and an attention module within the GNN architecture. Extensive experiments are conducted using the CALCE aging dataset and NASA battery dataset. The results substantiate the superiority of our method over existing state-of-the-art methods, with a mean absolute error (MAE) consistently averaging below 2%.