Online State Of Health Estimation Research Articles

Prediction of the health state of lithium-ion power batteries using machine learning

Faced with the increasingly urgent issues of energy depletion and environmental pollution, various countries are actively supporting the new energy vehicle industry, making electric vehicle technology a focus of widespread attention among researchers. The Battery Management System (BMS) plays a crucial role in overseeing the power battery's operational parameters, with the estimation of the State of Health (SOH) being a pivotal function. Accurate estimation of SOH can improve the utilization efficiency and endurance of power batteries, extend their service life, and help users achieve the best balance between system safety and economic benefits. Machine learning methods provide a new solution to the SOH estimation problem. Without analyzing the complex aging mechanisms inside the battery, online SOH estimation can be completed by learning from historical aging data. In this study, the lithium battery dataset provided by NASA is used and trained through multiple linear regression models and support vector machine models to predict the SOH of lithium-ion batteries. The results demonstrate the accuracy and reliability of both methods in predicting SOH. Simultaneously, the prediction results of different feature values are compared to obtain the most accurate combination of feature values.

An adaptive semi-supervised self-learning method for online state of health estimation of lithium-ion batteries

Accurate and precise online estimation of the state of health (SOH) is crucial when managing lithium-ion batteries. Most existing SOH estimation methods rely on supervised learning algorithms utilizing large amounts of labeled data. However, lithium-ion batteries are typically operated under dynamic conditions, including significant amounts of unlabeled charging or discharging data in online application scenarios. To fully utilize these data, we propose an adaptive semi-supervised self-learning teacher-student model (AS3LTSM) method for online SOH estimation. First, four physically interpretable health indicators (PIHIs) are extracted from the voltage and current data. The Pearson correlation coefficient (PCC) is used to assess significant associations between PIHIs and the SOH. Regressive and autoregressive long short-term memory (LSTM) models are selected as the teacher and student networks. Knowledge is transferred from the teacher to the student through pseudolabels, which guide the updating and evolution of the student network. Furthermore, a self-learning strategy and a retraining process for improving the long-term estimation accuracy are proposed. Two public datasets are used for comparison and ablation experiments. Experimental analysis validates the improved effectiveness and performance of the proposed method, with the RMSE and MAPE of the three experimental groups all within 1.3 % and 1.29 %, respectively.

A novel correlation-based approach for combined estimation of state of charge and state of health of lithium-ion batteries

State of charge(SOC) and state of health(SOH) estimation are two critical aspects for the secure operation and echelon utilisation of lithium-ion batteries. Considering the two states are interrelated, combined estimation methods are desired in practice. However, previous machine learning-based or traditional filter-based techniques have high computational requirements or suffer from convergence issues. To this end, the combined SOC and SOH estimation method of lithium-ion batteries using the novel correlation between SOC and SOH is proposed in this study. Specifically, a correlation between the capacity ratio for constant-current charging process and constant-voltage charging process and capacity loss is established. Extended Kalman filter is adopted to track SOC values at the end of constant current charging process instead of relying on the entire charging process during online application and SOH estimation can be accomplished in only one step based on the established correlation. The results show that the proposed combined SOC and SOH estimation method enables precise SOC estimation during both charging and discharging processes and achieves reliable estimation results during battery degradation. Additionally, the proposed method offers a swift means of estimating SOH with guaranteed estimation accuracy.

A Deep Learning Approach for Online State of Health Estimation of Lithium-Ion Batteries Using Partial Constant Current Charging Curves

A Deep Learning Approach for Online State of Health Estimation of Lithium-Ion Batteries Using Partial Constant Current Charging Curves

An online state-of-health estimation method for lithium-ion battery based on linear parameter-varying modeling framework

The accurate estimation of state-of-health (SOH) is crucial for ensuring the safe and reliable operation of lithium-ion battery systems. However, the intimate coupling between SOH and state-of-charge (SOC) is often overlooked in existing estimation methods, leading to inaccurate estimates. To address this, we propose a linear parameter-varying (LPV) battery model that captures both gradual capacity degradation and rapid dynamic changes. This model integrates traditional linear models with emerging nonlinear models, providing a comprehensive online SOH estimation framework that effectively separates the effects of SOC in the LPV model structure. The model parameters are identified using a subspace algorithm with accelerated aging data. The proposed method is validated by accelerated aging experiments on two sets of battery samples, one for model development and another for model validation. The experimental data show that the LPV battery model can achieve high SOH estimation accuracy, with an average error of 2.85 % and 5.51 % for SOH, and 0.63 % and 1.20 % for capacity, respectively. The method also shows the advantages of being easy to implement and highly generalizable, making it suitable for different battery types and application scenarios.

Deep Learning Powered Online Battery Health Estimation Considering Multitimescale Aging Dynamics and Partial Charging Information

Online accurate battery state-of-health (SOH) estimation is crucial for ensuring safe and reliable operations of electric vehicles (EVs). Yet, such estimation problem remains a challenge in reality due to complex battery degradation behaviors and dynamic EV operations. This paper proposes a novel deep learning-based framework, a bilateral branched Visual Transformer with Dilated Self-Attention (Bi-ViT-DSA), for online SOH estimation. The proposed framework considers partial charging segments during incomplete charging based on two mainstream charging modes, the multi-stage fast-charging and CCCV charging. To incorporate multi-timescale battery ageing dynamics into SOH estimation, a novel bi-party input structure is developed to convey both inner-cycle and intra-cycle degradation information from raw data. The proposed Bi-ViT-DSA is developed to learn multi-timescale high-level latent features from the bi-party input in parallel for SOH estimation. A dilated self-attention (DSA) mechanism is developed to reduce redundant operations in modeling. Computational studies are conducted on datasets of batteries under different chemistries and test conditions. Results validate the feasibility and robustness of the proposed method and its superior performance over a set of state-of-the-art benchmarks.

Online state of health estimation of lithium-ion batteries through subspace system identification methods

When Lithium-ion Batteries (LiBs) reach the end of their first life in electric vehicles (EVs), they can still be used in applications with lower power demands, a process known as second-life. However, to ensure that LiBs – or cells – removed from EVs operate safely, efficiently and reliably in a second application, several tests and procedures must be applied to study their internal conditions. Naturally, one of the most important parameters to be determined is the state of health (SoH). However, the available processes for determining the SoH of lithium-ion cells are limited by high costs, relatively long test times and the need for specific equipment, limiting the second-life market. Hence, this work proposes a methodology to estimate the SoH of lithium-ion cells, based on subspace system identification (SSI) methods, where the parameters estimated for the equivalent circuit model (ECM) of a given cell are associated with its SoH. To validate the proposed methodology, nine cell samples from the same manufacturer were considered, which were removed from heavy-duty EVs at the end of their first life. The obtained results showed that: (a) good approximations between the identified models and the actual cells were achieved, with root mean square error (RMSE) values as small as 1.32 mV; (b) SSI methods can be applied online, while the LiBs are still operating in the EV during their first life, eliminating the need of additional tests; and (c) there is a clear association between ECM parameters and the SoH, so it was possible to estimate the SoH of the samples with RMSE values varying from 2.11% to 3.34%. Therefore, the proposed methodology offers significant improvements when compared to the conventional capacity tests, including the possibility of estimating the SoH relatively fast, online and without the need for specific equipment.

Online state of health estimation for lithium-ion batteries based on gene expression programming

Lithium-ion battery is a kind of energy storage devices with complex internal reaction and many factors affecting its performance. Accurate prediction of its SOH (State of Health) is of great significance to prolong its service life and improve safety performance. However, the current prediction for SOH has the difficulties of selecting health factors and using data-driven methods with opaque mechanisms. In this paper, a data-driven model will be established with the capacity change during aging of lithium-ion batteries as a health indicator to realize precise prediction of capacity degradation. Firstly, the feature parameters were extracted and analyzed from the battery cyclic aging test dataset. Two in-situ nondestructive characterization methods, incremental capacity analysis curve and differential thermal voltammetry curve, were utilized to resolve the evolution paths of the feature parameters. After that, a data-driven battery capacity degradation estimation is realized based on the GEP (Gene Expression Programming) algorithm, the performance is compared with the existing vehicle-end and cloud-end models, and a higher-accuracy SOH stacking model is developed with an increase of less than 1 ms in computational time. The results indicated that the GEP model proposed in this paper has obvious advantages in terms of physical explain-ability, computational efficiency and robustness.

Li‐Ion Battery State of Health Estimation Using Hybrid Decision Tree Model Optimized by Bayesian Optimization

Accurately battery state of health (SOH) estimation in electric vehicles (EVs) is crucial for optimal performance and safety. This article presents an in‐depth investigation into the utilization of an optimized decision tree (DT) model for precise SOH estimation. Two aging features (AFs) are extracted through incremental capacity analysis from the partial charging process, and their efficacy is thoroughly validated on both battery cells and packs. To capture the relationship between AFs and capacity degradation, a robust DT model is established. Furthermore, Bayesian optimization is employed to optimize the model hyperparameters, enhancing the learning and generalization capabilities of the DT model. To validate the optimized DT model, a battery management system (BMS) is developed using a hybrid programming approach with LabVIEW and MATLAB, enabling online SOH estimation. The BMS undergoes rigorous evaluation using publicly available aging datasets of battery cells, as well as private datasets of battery packs. The results demonstrate that the optimized DT model outperforms traditional DT models, back propagation neural networks, and support vector machine approaches in accurately estimating battery SOH across diverse datasets. This research contributes to the advancement of BMS and holds profound implications for the efficient utilization of batteries in EVs.

Lithium-Ion Battery State of Health Estimation Using Simple Regression Model Based on Incremental Capacity Analysis Features

Batteries are the core component of electric vehicles (EVs) and energy storage systems (ESSs), being crucial technologies contributing to carbon neutrality, energy security, power system reliability, economic efficiency, etc. The effective operation of batteries requires precise knowledge of the state of health (SOH) of the battery. A lack of proper knowledge of SOH may lead to the improper use of severely aged batteries, which may result in degraded system performance or even a risk of failure. This makes it important to accurately estimate battery SOH using only operational data, and this is the main topic of this study. In this study, we propose a novel method for online SOH estimation for batteries featuring simple online computation and robustness against measurement anomalies while avoiding the need for full cycle discharging and charging operation data. Our proposed method is based on incremental capacity analysis (ICA) to extract battery aging feature parameters and regression using simple piecewise linear interpolation. Our proposed method is compared with back-propagation neural network (BPNN) regression, a popular method for SOH estimation, in case studies involving actual data from battery aging experiments under realistic discharging and temperature conditions. In terms of accuracy, our method is on par with BPNN results (about 5% maximum relative error), while the simplicity of our method leads to better computation efficiency and robustness against data anomalies.

Online State-of-Health Estimation for NMC Lithium-Ion Batteries Using an Observer Structure

State-of-health (SoH) estimation is one of the key tasks of a battery management system, (BMS) as battery aging results in capacity- and power fade that must be accounted for by the BMS to ensure safe operation over the battery’s lifetime. In this study, an online SoH estimator approach for NMC Li-ion batteries is presented which is suitable for implementation in a BMS. It is based on an observer structure in which the difference between a calculated and expected open-circuit voltage (OCV) is used for online SoH estimation. The estimator is parameterized and evaluated using real measurement data. The data were recorded for more than two years on an electrified bus fleet of 10 buses operated in a mild European climate and used regularly in the urban transport sector. Each bus is equipped with four NMC Li-ion batteries. Every battery has an energy of 30.6 kWh. Additionally, two full-capacity checkup measurements were performed for one of the operated batteries: one directly after production and one after two years of operation. Initial validation results demonstrated a SoH estimation accuracy of ±0.5% compared to the last checkup measurement.

Online State of Health Estimation with Deep Learning Frameworks Based on Short and Random Battery Charging Data Segments

Lithium-ion (Li-ion) batteries find wide application across various domains, ranging from portable electronics to electric vehicles (EVs). Reliable online estimation of the battery’s state of health (SOH) is crucial to ensure safe and economical operation of battery-powered devices. Here, we developed three deep learning models to investigate their potential for online SOH estimation using partial and random charging data segments (voltage and charging capacity). The models employed were developed from the feed-forward neural network (FNN), the convolutional neural network (CNN) and the long short-term memory (LSTM) neural network, respectively. We show that the proposed deep learning frameworks can provide flexible and reliable online SOH estimation. Particularly, the LSTM-based estimation model exhibits superior performance across the test set in both direct learning and transfer learning scenarios, while the CNN and FNN-based models show slightly diminished performance, especially in the complex transfer learning scenario. The LSTM-based model achieves a maximum estimation error of 1.53% and 2.19% in the direct learning and transfer learning scenarios, respectively, with an average error as low as 0.28% and 0.30%. Our work highlights the potential for conducting online SOH estimation throughout the entire life cycle of Li-ion batteries based on partial and random charging data segments.

State-of-health estimation for lithium-ion battery using model-based feature optimization and deep extreme learning machine

Accurate and reliable estimation of state of health (SOH) is essential for the safe and efficient operation of battery energy storages. However, random charging/discharging behaviors and ambient conditions complicate the online SOH estimation. Aiming at an accurate and robust online SOH estimation for lithium-ion batteries, this paper proposes a SOH estimation method using model-based feature optimization and improved machine learning. First, an empirical model-based voltage reconstruction method is proposed to reconstruct the voltage curve for solving disturbance-free different time (DT) curves under measurement noises and high C-rates and extract model-associated health features (HFs). Then, an optimal charging voltage window (CVW) determination method is proposed, which extracts the CVW-associated HFs by determining the optimal CVW from partial charging voltage ranges. Finally, a set of informative and multi-attribute HFs are extracted to train an improved deep extreme learning machine (DELM) mode for online SOH estimation. The proposed method is validated based on three datasets of batteries with different materials. The results demonstrate high accuracy and strong robustness to partial voltage range, noise corruption, and ambient temperature fluctuation.

Lithium-ion battery state of health estimation based on multi-source health indicators extraction and sparse Bayesian learning

A concise and accurate method for estimating the state of health (SOH) of lithium-ion batteries in the on-board energy management system is critical. However, SOH cannot be directly measured by on-board equipment. To improve the accuracy of SOH estimation for Li-ion batteries, this work proposes an SOH estimation model based on multi-source health indicators (HIs) extraction and sparse Bayesian learning. First, four direct HIs are extracted from the voltage and temperature curves of the batteries during charging and discharging, and two indirect HIs are extracted from the incremental capacity curves in combination with a Gaussian filtering algorithm. Then, the datasets are divided into three different training and test sets, which are used to simulate online SOH estimation under different situations. Finally, the six extracted HIs are selected using the Pearson correlation coefficient method, and the experiment is repeated for one of the situations using the three higher correlation features, and the results before and after selection are compared. The experimental results show that the proposed model can achieve satisfactory results in various simulated online estimation situations on the NASA and Oxford datasets.

Accurate state of health estimation for lithium-ion batteries under random charging scenarios

Accurate state of health (SOH) knowledge is critical for reliable operations of lithium-ion batteries. However, short-term random charging operations of lithium-ion batteries are not conducive to reliable SOH estimation. To solve it, a charging voltage prediction and machine learning based estimation are employed to supply precise estimation. Firstly, the correlated feature variables in terms of SOH are determined by analyzing the raw charging voltage distribution. Then, the wide range charging voltage is predicted via the constructed polynomials and short-term measures. Next, the extreme learning machine algorithm is employed to achieve online SOH estimation. Finally, the feasibility of the proposed voltage estimation method is verified at different aging cycles and different charging intervals, and the reliability of SOH estimation is investigated in the full lifetime range and at different state of charge (SOC) charging intervals. The experimental results manifest that the proposed method can supply reliable SOH estimation with the error of less than 2.02% based on only the short-term random charging scenarios, furnishing reliable and safe operation guideline for lithium-ion battery systems.

Linear Model for Online State of Health Estimation of Lithium-Ion Batteries Using Segmented Discharge Profiles

The pressing need to reduce CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> emissions has triggered the exponential growth of electric vehicles powered by lithium-ion batteries (LIBs). Accurate real-time State of Health (SOH) diagnosis of LIBs is paramount to ensure reliable operation of these batteries through their entire service life. This paper presents an intuitive multiple linear regression algorithm using short segment of voltage measurements from a battery’s constant current discharge profile to determine its SOH accurately with a Root Mean Square Error (RMSE) of less than 4%. The result from this work shows that voltage segments of just 0.02 V is sufficient to provide SOH prediction with less than 3% RMSE. However, the accuracy of the model is shown to be dependent on the specific voltage range of data used. To ensure robust SOH estimation and for practicality reasons, voltage segment of 0.1 V (about 13% SOC), within a range of 3.6 V to 3.9 V for nominal LIBs cell voltage (about 54 to 91% SOC), is recommended to be used for the SOH estimation. In this work, Nickel Cobalt Aluminum Oxides and Lithium Cobalt Oxide chemistry type are used for validation. The use of small voltage decay segments provides the capability for online SOH estimation.

Ensemble Learning and Voltage Reconstruction Based State of Health Estimation for Lithium-Ion Batteries With Twenty Random Samplings

Accurate estimation of state of health (SOH) is critical for the safe and efficient operation of lithium-ion batteries in electric transport tools. However, the random charge/discharge behaviors complicate online SOH estimation and discount estimation accuracy. To overcome this difficulty, this study presents an ensemble learning and voltage reconstruction-based SOH estimation framework through the incorporation of individual estimators and with the consideration of limited charging data. First, by analyzing more than 100 000 charging behaviors, the difficulty of feature extraction is addressed based on voltage distribution. Then, a voltage shape fitting method combing mechanistic and prognostic model is developed to reconstruct the constant current charge voltage, and the model parameters are identified by the moth-flame optimization algorithm. Next, the extreme learning machine and random forest are leveraged to estimate SOH preliminarily from the random discontinuous charging points with preferable diversity and high efficiency. On this basis, an induced ordered weighted averaging operator is exploited to efficiently integrate the individual learners and adaptively update the weight of each learner, thereby achieving better estimation than individual ones. The experimental results manifest that the SOH can be reliably estimated within an error of 3.42% using only 20 random samplings.

Artificial Intelligence-based health diagnostic of Lithium-ion battery leveraging transient stage of constant current and constant voltage charging

State of health (SOH) estimation is essential to the health diagnostic of lithium-ion battery. The data-driven approach with charging feature extraction is promising for online SOH estimation and has been widely explored over years. However, their deployment can be barriered by the lack of complete charging data in real-world applications. Motivated by this, this paper proposes an artificial intelligence-based SOH estimator using the transient phase between constant current (CC) and constant voltage (CV) charging, which is easily obtained in real-world charging scenarios. Specifically, a convolutional neural network (CNN) model is proposed to explain the relationship between the charging data and the SOH. Following this endeavor, the transfer learning is exploited for model mitigation and SOH estimation on different battery types, relying on much reduced amount of data for efficient CNN model re-training. The validation experiments are conducted based on the aging data obtained on LiNiCoAlO2 (NCA) and LiCoO2 (LCO) cells. Results suggest that the proposed method realizes accurate SOH estimation requiring only a short segment from the CC-CV transient phase, so that can meet a broad range of real-world charging scenarios. Moreover, the efficient model transfer promises expected performance with different battery types.The short version of the paper was presented at ICAE2021, Nov 29 - Dec 5, 2021. This paper is a substantial extension of the short version of the conference paper.

Online State-of-Health Estimation Method for Lithium-Ion Battery Based on CEEMDAN for Feature Analysis and RBF Neural Network

The state-of-health (SOH) estimation is of great importance in the battery management system. However, the accurate real-time SOH estimation is still a challenge because the capacity of the battery cannot be precisely monitored and measured in real-time. In response to this problem, this article proposes an online SOH monitoring method for lithium-ion battery (LIB) based on the charge–discharge feature of battery in real-time. First, the relationship between the charge–discharge feature and SOH through principal component analysis (PCA) is analyzed, and the appropriate feature which can precisely describe the battery aging process is then selected. Second, as to decompose and denoise the collected charge–discharge feature data, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is applied. The major trend obtained is predicted by logistic regression based on the sliding time window (LR-STW), and the minor fluctuation is predicted by Kalman filter (KF). Finally, an online SOH estimation method based on RBF neural network is proposed to establish the mapping relationship between feature and SOH. The proposed method was verified by the NASA experimental data. The experimental results show that the proposed method can accurately predict the LIB SOH in real-time, and the robustness of the proposed method is strengthened and promised.

Online state-of-health estimation algorithm for lithium-ion batteries in electric vehicles based on compression ratio of open circuit voltage

In electric vehicles (EVs), the online state-of-health (SOH) estimation is a challenging issue because of the unpredictable load behavior and the limited computational resources. In this study, a novel online SOH estimation algorithm for EVs is proposed based on the compression ratio of open circuit voltage (OCV)-to-charged capacity curve. The OCV-to-charged capacity curve has a clear correlation with the capacity degradation to be compressed with the same ratio of SOH, and this correlation is verified using five battery cells at the different aging states. The proposed algorithm estimates the degraded capacity at every sampling time during the driving operation through a first-order low-pass filter, which does not require complex mathematical tools and numerous offline data. For the robustness to the system noise in EVs, the estimated capacities are fed into a recursive average filter, and the SOH is updated to the output value of the average filter after sufficient driving operation. The experimental verifications were performed using five battery cells with different aging states, and these results represent the robustness and superiority of the proposed algorithm.