Lithium-ion Battery Capacity Estimation Research Articles

Adaptable capacity estimation of lithium-ion battery based on short-duration random constant-current charging voltages and convolutional neural networks

The accurate lithium-ion battery capacity estimation is vital for ensuring the safe and reliable operation of battery-powered systems. Existing data-driven methods heavily rely on fixed charging stages for feature extractions, posing significant limitations in real-world applications. This paper proposes an adaptable capacity estimation approach utilising short-duration random charging voltages during the constant-current charging stage and leveraging convolutional neural networks (CNNs). Based on the user-friendly “Vstart−tend” strategy, two health features including charging voltage and its increment are firstly extracted from random charging segments. Secondly, a feature evolution pattern analysis over the battery's lifespan is proposed to divide the charging voltage range for robust model development. An optimal combination of both the sampling interval and data length is determined for the feature extraction. Then, a two-dimensional CNN model is developed to effectively learn ageing-related knowledge from various random charging segments in a specific charging voltage range. The effectiveness of the proposed approach is ultimately verified using two distinct types of batteries across three operational temperatures. The results demonstrate that the proposed approach show much potential as a promising capacity estimation technique utilising a 600 s random charging segment sampled at a 20 s interval.

A transfer learning-based ensemble learning model for electric vehicles lithium-ion battery capacity estimation using electrochemical impedance spectroscopy

Lithium-ion battery capacity estimation is crucial to ensure the operational reliability and safety of electric vehicles. Electrochemical Impedance Spectroscopy (EIS) can provide rich physical degradation information of the battery, which makes the EIS-based data-driven method a promising solution for accurate battery capacity estimation. However, batteries tend to present diverse degradation patterns due to the diverse operating conditions and manufacturing, resulting in large capacity estimation errors in practical applications. Therefore, this paper proposes a transfer learning (TL) based ensemble learning (EL) method based on multi-layer perception for the battery capacity estimation. First, the influential impedance features are identified from the EIS using Spearman rank coefficient analysis. Second, three TL models are developed based on distinct TL strategies to enable sufficient knowledge transfer considering the different degradation pattern. Further, an EL method with Gaussian kernel-based weighted assignment technique is proposed to integrate the developed TL models and improve the estimation robustness. The performance of the developed model has been successfully validated only using the first 1/3 cycles of the historical data from target battery obtained under different operating conditions and batteries. The results reveal that the proposed method can largely suppress estimation errors, presenting more accurate capacity estimation compared to other methods with an RMSE lower than 1%.

Capacity estimation of lithium-ion battery with multi-task autoencoder and empirical mode decomposition

Capacity estimation of lithium-ion batteries is a commonly used method in health diagnosis and management. Its mainstream method involves using data-driven time series forecasting models to learn the patterns of changes in capacity. However, capacity regeneration poses a challenge for training time series forecasting models. Therefore, we propose a hybrid method that applies empirical mode decomposition and a multi-task autoencoder. In detail, empirical mode decomposition is applied to decompose the time series of capacity into intrinsic mode functions and a residual. Then, a multi-task autoencoder based on diagonal state space models is applied to estimate intrinsic mode functions while support vector regression is utilized for the residual. Experimental results show that the method outperforms seven baselines on three datasets, with an average root mean square error of 0.0103, 0.0111, and 0.0004. Furthermore, it is capable of performing an inference on the CPU in 3.57 ms with 0.69 MB of memory usage.

Intelligent Learning Method for Capacity Estimation of Lithium-Ion Batteries Based on Partial Charging Curves

Intelligent Learning Method for Capacity Estimation of Lithium-Ion Batteries Based on Partial Charging Curves

Capacity estimation of lithium-ion batteries with uncertainty quantification based on temporal convolutional network and Gaussian process regression

Reliable capacity estimation is crucial for safe operation of lithium-ion batteries (LIBs). This work combines the temporal convolutional network (TCN) and Gaussian process regression (GPR) to establish a novel probabilistic capacity estimation method. The proposed TCN-GPR method can not only provide accurate capacity estimation but also quantify the uncertainty of the estimation. Besides, the TCN-GPR method can automatically extract degradation features from partial charging segments, overcoming the limitations of manual experience. In addition, the TCN-GPR method can be applied to different types of LIBs through transfer learning using only a small amount of training data. For validation, the Oxford battery dataset is used to demonstrate the accuracy and robustness of the TCN-GPR method, where a mean absolute percentage error (MAPE) of less than 0.3% can be achieved with only a 15-min partial charging segment. Furthermore, our own experimental dataset is used to demonstrate the generalization ability of the TCN-GPR method through transfer learning, where a MAPE of less than 0.7% can be achieved by using only one battery cell as the training sample.

Online capacity estimation of lithium-ion batteries based on convolutional self-attention

ABSTRACT It is of great significance to improve the safety, efficiency, and economy of lithium-ion batteries by improving the capacity estimation accuracy of lithium-ion batteries. In this paper, feature extraction and correlation analysis are carried out on the data of lithium-ion battery charging process, and the voltage curve of constant current charging stage is extracted. The difference characteristics between each cycle are used to describe the battery capacity, and these statistical characteristics are proved to be highly correlated with the battery capacity. Furthermore, an online estimation model of battery capacity based on convolution self-attention is established, and the above characteristics of constant current charging process and the battery capacity of the latest cycle are fused as input vectors of the model to realize online estimation of battery capacity. Finally, an open data set is used to verify the model experiment. The results show that the prediction error of MAE is 0.17% and that of RMSE is 0.22%.

A novel variable activation function-long short-term memory neural network for high-precision lithium-ion battery capacity estimation

A novel variable activation function-long short-term memory neural network for high-precision lithium-ion battery capacity estimation

Semi-supervised adversarial deep learning for capacity estimation of battery energy storage systems

Battery energy storage systems (BESS) play a pivotal role in energy management, and the precise estimation of battery capacity is crucial for optimizing their performance and ensuring reliable power supply. Deep learning methodologies applied to battery capacity estimation have exhibited exemplary performance. However, deep learning methods necessitate supervised training with a significant volume of labeled data, presenting challenges for data collection in industrial scenarios. Moreover, a diverse range of battery types in industrial settings makes it difficult to develop capacity estimation models for different types of batteries from scratch. To address these issues, a semi-supervised adversarial deep learning (SADL) method is proposed for lithium-ion battery capacity estimation. Initially, a subset of labeled lithium-ion battery data, coupled with a subset of unlabeled data, is collected. Voltage and current data are then transformed into capacity increment features. Subsequently, an adversarial training strategy is employed, subjecting labeled and unlabeled data to adversarial training to enhance the performance of SADL. Finally, the effectiveness of the SADL method in estimating the capacity of other lithium-ion batteries is analysed. Experimental results demonstrate that the SADL method accurately estimates the capacity of various battery types, showcasing an RMSE error of approximately 2%, surpassing the performance of other methods. The proposed SADL method emerges as a promising solution for the precise estimation of lithium-ion battery capacity in BESS.

Investigation into Impedance Measurements for Rapid Capacity Estimation of Lithium-ion Batteries in Electric Vehicles

With the dramatic increase in electric vehicles (EVs) globally, the demand for lithium-ion batteries has grown dramatically, resulting in many batteries being retired in the future. Developing a rapid and robust capacity estimation method is a challenging work to recognize the battery ageing level on service and provide regroup strategy of the retied batteries in secondary use. There are still limitations on the current rapid battery capacity estimation methods, such as direct current internal resistance (DCIR) and electrochemical impedance spectroscopy (EIS), in terms of efficiency and robustness. To address the challenges, this paper proposes an improved version of DCIR, named pulse impedance technique (PIT), for rapid battery capacity estimation with more robustness. First, PIT is carried out based on the transient current excitation and dynamic voltage measurement using the high sampling frequency, in which the coherence analysis is used to guide the selection of a reliable frequency band. The battery impedance can be extracted in a wide range of frequency bands compared to the traditional DCIR method, which obtains more information on the battery capacity evaluation. Second, various statistical variables are used to extract ageing features and Pearson correlation analysis is applied to determine the highly correlated features. Then a linear regression model is developed to map the relationship between extracted features and battery capacity. To validate the performance of the proposed method, the experimental system is designed to conduct comparative studies between PIT and EIS based on the two 18650 batteries connected in series. The results reveal that the proposed PIT can provide comparative indicators to EIS, which contributes higher estimation accuracy of the proposed PIT method than EIS technology with lower time and cost. Conflict of Interest Statement The authors declare no conflicts of interest.

Comprehensive Analysis of Charging Profile Dynamics for Lithium-ion Battery Capacity Estimation

Comprehensive Analysis of Charging Profile Dynamics for Lithium-ion Battery Capacity Estimation

Lithium-ion battery capacity estimation based on fragment charging data using deep residual shrinkage networks and uncertainty evaluation

Accurate and reliable capacity estimation is crucial for lithium-ion batteries to operate safely and stably. However, the extraction steps of health indicators (HIs) limit the feasibility and applicability of data-driven methods. This study proposes a novel estimation framework using deep residual shrinkage network (DRSN) and uncertainty evaluation to estimate the lithium-ion battery capacity directly; model inputs are only random fragment charging data. Results on three datasets confirm that accurate capacity estimation is achieved by DRSN through integrated attention mechanisms and soft thresholding (the mean absolute percentage error is below 2 %). Meanwhile, the inherent noise sensitivity in data-driven methods is alleviated. Additionally, uncertainty evaluation implemented by Bayesian neural networks provides valuable metrics for the reliability of estimated results from different voltage ranges. The significant correlation between uncertainty and error proves the potential of uncertainty to assist battery management systems for control and decision-making. The experimental study demonstrates the high accuracy, adaptability, and robustness of the proposed framework, especially in coping with the frequent occurrence of random incomplete charging scenarios.

Capacity estimation for lithium-ion batteries with short-time working condition in specific voltage ranges

The data-driven methods for estimating battery capacity utilize complete charging/discharging cycle data and extract characteristic parameters to provide accurate battery capacity, but the method is consuming long time, limiting application in reality. In order to overcome this limitation, we propose a short-time equivalent power working condition as the battery aging condition method and obtain the training dataset, and introduce the K-means clustering algorithm to determine high-middle-low three specific voltage segments of the discharge curve. Eight feature parameters are extracted from the battery-specific voltage segments and dimension-reduction using principal component analysis (PCA) method for capacity estimation. Convolutional neural network- long short-term memory network-attention (CNN-LSTM-ATT) is used to establish capacity estimation model in specific voltage, and its performance is compared with other machine learning algorithms. Finally, different SOH batteries are selected for short-time equivalent working condition discharges to verify the validity of the estimation model. This work emphasizes the application potential of combining short-time equivalent power work condition with deep learning algorithms for capacity estimation of lithium-ion batteries.

Enhanced robust capacity estimation of lithium-ion batteries with unlabeled dataset and semi-supervised machine learning

The capacity estimation is a crucial task in battery health and safety management. The majority existing capacity estimation methods heavily rely on supervised learning with accurately labelled dataset collected at room temperature. However, the unlabeled dataset and the influence of ambient temperature on lithium-ion batteries degradation are rarely considered in the existing works. To address these issues, semi-supervised machine learning based noise-immune capacity estimation approach is proposed by utilizing both labelled and unlabeled datasets. By considering both the deployment and the accuracy, the optimal health indicator is extracted based on detailed analysis of limited labelled dataset. Then, the semi-supervised machine learning is proposed to improve the capacity estimation by using abundant unlabeled dataset. Specifically, low-complexity network with different training algorithms are employed to map the nonlinear relationship between the feature and capacity with the total loss of labelled and unlabeled datasets. By further exploiting the unlabeled dataset, the maximum errors of capacity estimation at three different temperatures are improved by 85%, 13%, and 48% with strong robustness, respectively. The proposed approach shows superior performance to the state-of-the-art supervised methods. The encouraged performance indicates the effectiveness of using large unlabeled battery dataset to improve the capacity estimation in real-world applications.

A fast data-driven battery capacity estimation method under non-constant current charging and variable temperature

Data-driven methods have been widely used in capacity estimation of lithium-ion batteries. Non-constant current charging and variable-temperature operating scenarios are inevitable in real applications. However, existing classical constant current charging based capacity estimation methods may not be applied to such scenarios. To this end, a data-driven method adapted to non-constant current charging is proposed in this work. A single health feature collected within 3 min is combined with Gaussian process regression (GPR) to achieve fast and transferable battery capacity estimation. To validate the effectiveness of the proposed method, ten batteries are performed for aging tests using a non-constant current charging protocol. The validation results show that the proposed method's average capacity estimation error (mean absolute percentage error) is only 0.65 %. The estimation error is reduced by more than 30 % for the new temperature condition. Compared with the three existing classical methods, the capacity estimation accuracy of the proposed method is improved by 69 %, 82 %, and 68 %, respectively. Additional experiments demonstrate the generality of the proposed approach for different battery chemistries and charging protocols. This work opens the door to battery capacity estimation for non-constant current charging scenarios.

Capacity estimation of Li-ion battery based on transformer-adversarial discriminative domain adaptation

Lithium-ion batteries are widely used in various electronic devices as well as electric vehicles, and accurate estimation of the battery capacity is important to ensure safe and reliable operation of the system. However, in practice, the complex working conditions and the limitation of the number of charge/discharge cycles lead to insufficient historical data and inaccurate capacity estimation. In order to improve the adaptability as well as accuracy under different operating conditions, this paper proposes a lithium-ion battery capacity estimation model based on Transformer-Adversarial Discriminative Domain Adaptation (T-ADDA). The model takes charging voltage, charging current, and charging temperature as inputs and uses a transformer network to extract the time series features from the data. Then, adversarial domain adaptation is trained on the source and target domain data by the domain discrimination network of the ADDA model so as to find the domain invariant features between the source and target domains. Finally, the regression network of ADDA is used to achieve cross-domain prediction for the target domain data. The experimental results show that the T-ADDA model can accurately achieve cross-domain prediction and that the average error of prediction under different operating conditions is only 3.9225%. Therefore, the T-ADDA model has good adaptability and accuracy, and it can significantly improve the performance of lithium-ion battery capacity estimation.

Capacity estimation of lithium-ion batteries based on adaptive empirical wavelet transform and long short-term memory neural network

To ensure the durability and safety of electric vehicles (EVs), it is vital to monitor the capacity deterioration of lithium-ion batteries (LIBs). However, due to complex physicochemical interactions and temperature effects, the capacity of LIBs cannot be directly measured. This work presents a novel generalized approach for estimating the capacity of LIBs based on the adaptive empirical wavelet transform (EWT) and long-short-term memory neural network (LSTM). The adaptive EWT is a potent tool for analyzing the charging-discharging terminal voltage signal with non-stationary and transient phenomena of the LIBs in the time-frequency domain. Specifically, the measured voltage signals of the LIBs are decomposed into nine multi-resolution modes to display the high and low-frequency components. Then, 13 statistical features are extracted to investigate the correlation between the capacity degradation and the extracted features. Afterwards, the LSTM model is developed to estimate the capacity of the LIBs. The proposed approach has been validated using two datasets: (1) NASA's randomized dataset with 24 LIBs cycled under generally varying operational conditions and (2) Stanford University's dataset with 10 LIBs cycled with EV discharge current profile. Compared with the state-of-the-art, the proposed method accurately estimates LIB capacity with an average root mean square error of 1.26 % and a maximum error of 2.74 %.

An improved sliding window - long short-term memory modeling method for real-world capacity estimation of lithium-ion batteries considering strong random charging characteristics

Capacity estimation plays a significant role in ensuring safe and acceptable energy delivery, especially under real-time complex working conditions for whole-life-cycle lithium-ion batteries. For high-precision and robust capacity estimation, an improved sliding window-long short-term memory (SW-LSTM) modeling method is proposed by introducing multiple time-scale charging characteristic factors. The optimized feature information set is extracted by constructing an optimized differential integration-moving average autoregressive (DI-MAA) model, which is introduced as the input matrices of the whole-life-cycle capacity estimation model. With the constructed DI-MAA model, the relevant features are effectively extracted, overcoming the data limitation problem of the long-term dependence capacity estimation. For the experimental test, the maximum capacity estimation error is 3.56 %, and the average relative error is 0.032 under the complex Beijing bus dynamic stress test working condition. The proposed SW-LSTM estimation model with optimized DI-MAA-based data pre-processing treatment has high stability and robust advantages, serving an effective safety assurance for lithium-ion batteries with real-world complex working condition adaptation advantages.

Data-driven lithium-ion batteries capacity estimation based on deep transfer learning using partial segment of charging/discharging data

Accurate estimation of lithium-ion battery capacity is crucial for ensuring its safety and reliability. While data-driven modelling is a common approach for capacity estimation, obtaining cycling data during charging/discharging processes can be challenging. Collecting cycling data under various charging/discharging protocols is often unrealistic, and the collected data can be fragmented due to the random nature of working conditions in practice. To address these issues, we propose a deep transfer learning method that uses partial segments of charging/discharging data for battery capacity estimation. The proposed method utilizes capacity increment features of partial charging/discharging segments that is designed to satisfy practical scenarios. A deep transfer convolutional neural network (DTCNN) is trained with both source and target data, and a fine-tuning strategy is employed to effectively eliminate distribution discrepancies between different battery types or charging/discharging protocols, leading the improved estimation accuracy. Experimental results demonstrate that the proposed method accurately estimates the lithium-ion battery capacity, with values of RMSE, MAPE, and MD-MAPE of only 0.0220, 0.0247, and 0.0194, respectively, when using partial segments. These results highlight the promising prospects of the proposed method for lithium-ion battery capacity estimation.

A Flexible deep convolutional neural network coupled with progressive training framework for online capacity estimation of lithium-ion batteries

Machine learning-based methods have shown great application prospects on the capacity estimation of lithium-ion battery. However, most of extant approaches require complete charging and discharging profiles that rarely occurs in practical applications to extract health features matched by the cycling capacity. To address this limitation, this paper directly intercepted partial charging voltage segments from raw charging curves, evading complicated manual features extraction. A progressive training framework is proposed and integrated with the deep convolutional neural network to build the online capacity estimator. Three battery aging datasets involving 28 cells cyclic aging data are exploited for model training and validation. Thereinto, the source dataset and target dataset-1 are used to train the model parameters progressively, and the target dataset-2 is applied to validate model performance. Experimental results show that the deep convolutional neural network coupled with progressive training framework can significantly enhance the capacity estimation accuracy and reduce retraining time compared with the ordinary convolutional neural network. The comparison results of different machine learning algorithms demonstrate that the proposed method possesses high estimation precision with a maximum relative error of only 3.4% on the target dataset, which can be easily extended to different types of battery capacity estimation.

Capacity estimation of lithium-ion batteries based on data aggregation and feature fusion via graph neural network

Lithium-ion batteries in electrical devices face inevitable degradation along with the long-term usage. The accompanying battery capacity estimation is crucial for battery health management. However, the hand-crafted feature engineering in traditional methods and complicated network design followed by the laborious trial in data-driven methods hinder efficient capacity estimation. In this work, the battery measurements from different sensors are organized as the graph structure and comprehensively utilized based on graph neural network. The feature fusion is further designed to enhance the network capacity. The specific data aggregation and feature fusion operations are selected by neural architecture search, which relieves the network design and increases the adaptability. Two public datasets are adopted to verify the effectiveness of the proposed scheme. Additional discussions are conducted to emphasize the capability of the graph neural network and the necessity of architecture searching. The comparison analysis and the performance under noisy environment further demonstrate the superiority of proposed scheme.