Li-ion Battery State Of Charge Research Articles

Li-Ion Battery State of Charge Prediction for Electric Vehicles Based on Improved Regularized Extreme Learning Machine

Battery state of charge prediction is one of the most essential state quantities of a battery management system. It is a prerequisite for the operation of a battery management system, but it becomes difficult to make an exact prediction of its state due to its characteristics, which cannot be measured directly. For the exact assessment of the Li-ion battery state of charge, the research proposes an extreme learning machine algorithm based on the alternating factor multiplier method with improved regularization. This method constructs a suitable online Li-ion battery state of charge prediction model using the alternating factor multiplier method in gradient form. The experiment demonstrates that the algorithm in the study has a reduction in the number of nodes in the implicit layer relative to the traditional extreme learning machine algorithm. The error fluctuations of the algorithm under two different excitation functions range from [−0.005, 0.005] and [0.082, 0.265]; The root mean square error of the data set in which the algorithm performs well is 1.9516 and 0.6157, respectively. The real simulation scenario created the predicted values of the state of charge in the realistic simulation scenario that fit the real value curve by 99.99%. The average and maximum errors of the proposed state of charge prediction model are the smallest compared to the long and short-term memory networks and gated cyclic units, 0.58% and 2.97%, respectively. The experiment demonstrates that the presented algorithm can reduce the computational burden while guaranteeing the state of charge model prediction.

A Novel Method for SoC Estimation of Lithium-Ion Batteries Based on Kalman Filter in Electric Vehicle

In recent years, the energy crisis has become more and more serious. Li-ion batteries are used in grids because of their benefits such as contributing to the intermittent generation of renewable energy sources and stabilizing the grid. In addition, li-ion batteries are widely used in electric vehicles due to their long cycle life and high energy density. Li-ion battery state of charge (SoC) is an important indicator for safety. Therefore, the SoC estimation of li-ion batteries is important. Today, there are different methods to determine the state of the SoC in many applications. The traditional estimation method, the ampere-hour integration method and the coulomb counting method, has a cumulative error and cannot achieve good results in a working environment with Gaussian noise. For this purpose, in this study, firstly, the Thevenin equivalent model was created for battery SOC estimation, and then the Kalman filter algorithm was applied. Thus, the estimation error caused by Gaussian noise is eliminated. SoC estimation was simulated for the battery model created in the MATLAB/Simulink program using this method. Using these simulation results, the charge/discharge characteristics of the battery were obtained. However, the SoC estimation has been made for the charging and discharging processes of the battery. In the simulation, the charge value was recorded for 6 hours. The data recorded every 10 minutes gave results very close to the true value.

Online fusion estimation method for state of charge and state of health in lithium battery storage systems

To ensure the safe and reliable operation of Li-ion battery energy storage systems, it is important to diagnose the operational status and aging degree of the batteries. In this study, an online fusion estimation method based on back propagation neural network and genetic algorithm (BP-GA) is used for estimating the state of charge (SoC) and state of health (SoH) of Li-ion batteries. First, the effective features of SoC and SoH of Li-ion batteries during charging and discharging are analyzed, and the relevant features are extracted. Subsequently, a conventional back propagation neural network (BPNN) for SoC and SoH estimation is described. The extracted feature quantities are then used as inputs to the neural network for the estimation of the SoC and SoH of Li-ion batteries. A comparison of the SoC and SoH estimation results using, respectively, a single BPNN and a fusion algorithm using a genetic algorithm (GA) to optimize the BPNN shows that the accuracy and efficiency of the estimation results using BP-GA are significantly higher than that of the BPNN. The experimental component of this paper is a comparison experiment using the Panasonic 18650PF and the real-world data from NASA. The results show that the feature extraction operation is relatively easy to perform and combined with the neural network method used in this paper can have good accuracy, reduce the complexity of the algorithm, and improve the detection efficiency.

Analysis of a Li-ion battery state of charge by artificial neural network

The state of charge (SOC) is a battery residual capacity crucial assessment metric. The need for a precise SOC estimate is very important to ensure the safe functioning of a Li-ion battery and to prevent overload and over-depletion. However, the renewable energy-based standalone application has become a key problem to determine the exact capacity of SOC of the Li-ion battery. To estimate the capacity over time, the battery management system calculates the SOC of a Li-ion battery. This allows for the implementation of intelligent control systems. This paper presents an enhanced radial basis function (RBF) of the SOC battery estimate following the limits and weaknesses of the back propagation (BP) neural network (NN) in estimating battery SOC, such as sluggish convergence speed, poor generalization and can increase the accuracy of the network but it takes time to iterate. Train the enhanced RBF with experimental data in real-time. The trained NN of SOC is compared to actual values and the MATLAB is used to simulate the method to evaluate its accuracy.

Lithium-Ion Battery State-of-Charge Estimation Using Electrochemical Model with Sensitive Parameters Adjustment

State-of-charge (SOC) estimation of lithium-ion (Li-ion) batteries with good accuracy is of critical importance for battery management systems. For the model-based methods, the electrochemical model has been widely used due to its accuracy and ability to describe the internal behaviors of the battery. However, the uncertainty of parameters and the lack of correction from voltage also induce errors during long-time calculation. This paper proposes a particle filter (PF) based method to estimate Li-ion batteries’ SOC using electrochemical model, with sensitive parameter identification achieved using the particle swarm optimization (PSO) algorithm. First, a single particle model with electrolyte dynamics (SPME) is used in this work to reduce the computational burden of the battery electrochemical model, whose sensitive parameters are selected through the elementary effect test. Then, the representative sensitive parameters, which are difficult to measure directly, are adjusted by PSO for a high efficiency. Finally, a model-based SOC estimation framework is constructed with PF to achieve accurate Li-ion battery SOC. Compared with extended Kalman filter and equivalent circuit model, the proposed method shows high accuracy under three different driving cycles.

State of charge effects on active material elemental composition changes between pre-thermal-runaway and post-failure states for 8-1-1 nickel-manganese-cobalt 18650 cells

Lithium-ion battery failures have presented challenges to fire forensic investigators in determining in what scenarios they may have been able to initiate a fire, and what post-failure signatures exist to determine if the battery was more likely the cause or a victim of the fire. A battery's ability to initiate a fire depends on its state-of-charge (SOC) which influences a failed cell's energy output and capacity to ignite fuel contents. Knowing a failed battery's pre-failure SOC is thus of great interest in battery failure forensics in helping analyze whether it is the cause or a victim of a fire. To determine the pre-failure SOC of burned batteries, a simple and repeatable technique is needed to analyze burned batteries and provide insight on the possible pre-failure SOC. This paper introduces a technique that deduces the pre-failure SOC based on the elemental composition of 8-1-1 NMC Li-ion batteries before and after thermal runaway at different SOCs. The composition-SOC correlation was determined by (1) acquiring a baseline composition from the Nickel-Manganese-Cobalt (NMC) cathode material in a pristine 18650 cell, (2) triggering thermal runaway of 18650 cells (by overheating) at different SOCs and collecting burned electrodes, (3) performing chemical analysis using scanning electron microscopy (SEM) and energy dispersive x-ray spectroscopy (EDS) to obtain post-failure composition, and (4) examining elemental composition variation with respect to SOC. In addition to characterizing the burned contents inside, the ejected (gas and solid) contents during thermal runaway were also collected in a sealed pressure vessel chamber and analyzed to explain the mechanistic background behind the observed variation. The ejected solids were analyzed using SEM-EDS, and the ejected gases were analyzed using gas chromatography (GC) and gas detection tubes. The results show that as the SOC increases, the Ni/Mn and Co/Mn ratios after thermal runaway decrease from the atomic ratios in the pristine condition. Test repetition at each SOC produced an uncertainty in the Ni/Mn ratios of less than 10 % in the 30–70 % SOC tests and less than 20 % in the 100 % SOC tests for externally heated cells, and less than 10 % in the 30 % SOC tests and less than 25 % in the 50–100 % SOC tests for internally heated cells. Ni/Co ratio does not vary significantly with respect to the SOC, suggesting that nickel and cobalt are lost from the electrode at a faster rate than manganese, with nickel being lost slightly faster than cobalt. The pressure vessel test conducted with 100 % SOC cells has also been found to contain nickel carbonyl in the inert test chamber after thermal runaway, which may be a product of nickel reactions in the electrode. Overall, this paper introduces a practical approach to determine a NMC (8,1,1) Li-ion battery's SOC according to the elemental composition collected from its debris after a failure event. The approach is based on the correlation between SOC and Ni-Mn-Co ratios and supported by the theoretical reaction kinetics of NMC cathode materials during thermal runaway, as evidenced by the identified Ni-containing products.

Battery management solutions for li-ion batteries based on artificial intelligence

The automobile industry is currently undergoing a paradigm change from conventional, diesel, and gasoline-powered vehicles to hybrid and electric vehicles of the second generation. Lithium-ion (Li-ion) batteries have sparked the automotive industry’s interest for quite some time. One of the most crucial components of an electric car is the battery management system (BMS). Since the battery pack is an electric vehicle's most significant and expensive component, it must be carefully monitored and controlled. The precise measurement and calculation of the many states of a Li-ion battery's cells, such as the State of Health (SOH) and State of Charge (SOC) is a difficult procedure as they cannot be monitored directly. This paper examines various methodologies and approaches for estimating the SOC and SOH of Li-ion batteries using Artificial Intelligent methods. Six machine learning algorithms are intensively utilized to investigate the Li-ion battery state estimation. The employed methods are linear, random forest, gradient boost, light gradient boosting (light-GBM), extreme gradient boosting (XGB), and support vector machine (SVM) regressors. In comparison to all other models employed in this study, the discharge prediction made using random forest exhibits significantly greater performance at a low loss of accuracy. For instance, with the highest R2-score of 0.999, the random forest regressor achieves only 0.0035, 0.0013, and 0.0097 for mean and median absolute error, and root means squared error (RMSE), respectively. We showed that the state estimation of Li-ion batteries can be precisely predicted using AI methods, which can be combined with a battery management system to improve electric vehicle performance.

The Scheme for SOC Estimation of Lithium-ion Batteries based on EQ-OCV-Ah-EKF

The state of charge (SOC) of a battery is one of the most important indicators to monitor in a Battery Management System (BMS). It is very important to accurately and quickly predict the state of battery charge for the safe driving of electric vehicles and the efficient utilization of battery equipment. However, several factors, such as the working environment and the aging degree of the battery, can have a coupling effect on the internal state of the battery. This leads it difficult to accurately predict the SOC of a battery. For this purpose, this paper proposes an online estimation scheme for the SOC of Li-ion batteries with an equivalent circuit model and an extended Kalman filtering algorithm. Firstly, a batch of ternary Li-ion batteries are charged and discharged at constant current with 1C current condition under constant temperature at 25°C using the hybrid power pulse (HPPC) test method, and the collected data are pre-processed. Then, the parameters of the second-order Thevenin circuit model were fitted by combining the Genetic Algorithm (GA) and Levenberg-Marquardt algorithm (LM) and verified by the MATLAB program. Finally, the Extended Kalman algorithm (EKF) is leveraged to further improve the accuracy of prediction results. The proposed scheme also has some reference significance for the identification of equivalent circuit model parameters of lithium-ion batteries and the rapid estimation of battery charge states.

Joint Prediction of Li-ion Battery State of Charge and State of Health Based on the DRSN-CW-LSTM Model

Joint Prediction of Li-ion Battery State of Charge and State of Health Based on the DRSN-CW-LSTM Model

Estimating battery state-of-charge with a few target training data by meta-learning

This study proposes a meta-learning battery state-of-charge (SOC) estimation approach to reduce the amount of target battery data required for training a Li-ion battery SOC estimation by applying deep-learning. The proposed approach reduces the training data required for the target battery by improving the pre-training performance. Multiple other batteries are used as pre-training data to allow generalization to the target battery. The performance of meta-learning SOC estimation is compared with that of transfer learning using one battery data for pre-training. The meta-learning performance does not depend on the similarity between the pre-training data and target battery data. The proposed meta-learning SOC estimation accuracy is verified based on battery test data from various driving cycles (US06, urban dynamometer driving schedule (UDDS), and LA92), to reflect actual electric vehicle driving patterns. Using only a small amount of target battery data and nine gradient steps, the proposed meta-learning SOC estimation algorithm achieves a mean squared error (MSE) of 0.0176% and a mean absolute error (MAE) of 1.0075%. These results show that the proposed method can adapt more quickly than transfer learning (with SOC estimation errors of 3.1378% in terms of the MSE and 15.1327% in terms of the MAE under the same conditions).

Deep Learning in the State of Charge Estimation for Li-Ion Batteries of Electric Vehicles: A Review

As one of the critical state parameters of the battery management system, the state of charge (SOC) of lithium batteries can provide an essential reference for battery safety management, charge/discharge control, and the energy management of electric vehicles (EVs). To analyze the application of deep learning in electric vehicles’ power battery SOC estimation, this study reviewed the technical process, common public datasets, and the neural networks used, as well as the structural characteristics and advantages and disadvantages of lithium battery SOC estimation in deep learning methods. First, the specific technical processes of the deep learning method for SOC estimation were analyzed, including data collection, data preprocessing, feature engineering, model training, and model evaluation. Second, the current commonly and publicly used lithium battery dataset was summarized. Then, the input variables, data sets, errors, and advantages and disadvantages of three types of deep learning methods were obtained using the structure of the neural network used for training as the classification criterion; further, the selection of the deep learning structure for SOC estimation was discussed. Finally, the challenges and future development directions of lithium battery SOC estimation using the deep learning method were explained. Over all, this review provides insights into deep learning for EVs’ Li-ion battery SOC estimation in the future.

Ultrasonic nondestructive diagnosis of lithium-ion batteries with multiple frequencies

Accurately estimating the state of charge (SoC) in battery management systems (BMSs) requires the measurement of numerous parameters and advanced algorithms. This work studies multifrequency ultrasonic waves to estimate the SoC of Li-ion batteries by sensing the material changes during charge/discharge. A pouch-type LiNi0.6Mn0.2Co0.2O2 (NMC622)||graphite battery cell is designed and fabricated with a capacity of 2.4 Ah. Different ultrasonic testing setups are explored to determine the optimal testing parameters for the battery. An ultrasonic monitoring system is developed to monitor the battery during charge/discharge at 750 kHz, 1 MHz, and 1.5 MHz. Signal processing algorithms are proposed for extracting three ultrasonic features—amplitude, wave velocity, and attenuation. In a three-cycle test, the amplitude histories do not show clear correlations with the SoC. The wave velocities of all three frequencies have an approximately linear relationship with the SoC, which can be used for SoC estimation. Hysteresis behavior is observed for the wave velocity in terms of a larger slope in the discharge process and velocity drop after a close charge/discharge cycle. The wave attenuation is able to capture the material phase transitions during charge/discharge.

Online Estimations of Li-Ion Battery SOC and SOH Applicable to Partial Charge/Discharge

Estimating the state of health (SOH) and state of charge (SOC) of lithium-ion batteries is crucial for increasing the battery lifetime and performance. Many estimation methods are offline and require large datasets for training. The majority of online estimation methods either take too much time or need a full discharge or charge cycle. In this article, a fast online SOH estimation method that can work with partial charge/discharge is introduced. Only two consecutive partial discharge intervals are used to estimate the battery equivalent circuit model parameters and the open-circuit voltage (OCV). By comparing the estimated OCV curve at each interval with a reference or datasheet OCV curve, the battery capacity and, therefore, its SOH and SOC are accurately estimated. It is shown that updating the OCV reference curve based on temperature readings will provide more accurate results. NASA degradation dataset is used to validate the proposed method and the average reported root-mean-square error is below 1% for SOH and 1.07% for SOC.

A Novel Data-Driven Estimation Method for State-of-Charge Estimation of Li-Ion Batteries

With the increasing proportion of Li-ion batteries in energy structures, studies on the estimation of the state of charge (SOC) of Li-ion batteries, which can effectively ensure the safety and stability of Li-ion batteries, have gained much attention. In this paper, a new data-driven method named the probabilistic threshold compensation fuzzy neural network (PTCFNN) is proposed to estimate the SOC of Li-ion batteries. Compared with other traditional methods that need to build complex battery models, the PTCFNN only needs data learning to obtain nonlinear mapping relationships inside Li-ion batteries. In order to avoid the local optimal value problem of traditional BP neural networks and the fixed reasoning mechanism of traditional fuzzy neural networks, the PTCFNN combines the advantages of a probabilistic fuzzy neural network and a compensation fuzzy neural network so as to improve the learning convergence speed and optimize the fuzzy reasoning mechanism. Finally, in order to verify the estimation performance of the PTCFNN, a 18650-20R Li-ion battery was used to carry out the estimation test. The results show that the mean absolute error and mean square error are very small under the conditions of a low-current test and dynamic-current test, and the overall estimation error is less than 1%, which further indicates that this method has good estimation ability.

Electrochemical impedance correlation analysis for the estimation of Li-ion battery state of charge, state of health and internal temperature

Electrochemical impedance spectroscopy (EIS) is an effective characterization tool for a multitude of battery states including state of charge (SoC), state of health (SoH) and internal temperature (IT). The intrinsic relationship between equivalent circuit elements and components of an impedance spectra (frequency, real, imaginary and phase) could be exploited to estimate the battery state at a given point of time without the need of continuous historical tracking information. Identification and analysis of battery state sensitive impedance variables is paramount for the development of any impedance-based battery management system (BMS). In this paper, correlation analysis between equivalent circuit elements and impedance spectra of multiple commercial Li-ion polymer batteries at varying SoC, SoH and IT levels was performed to identify and quantify the degree of dependence. Curve fitting techniques were used to fit the measured Impedance spectra on to an equivalent circuit model (ECM) to extract the circuit elements. Pearson's r correlation matrix was employed for quantifying the degree of correlation between each impedance variable and state parameter. Optimal impedance variables that demonstrated high dependence with SoC, SoH and IT are then proposed in this paper. Knowledge of this information is of high value to develop a direct impedance-based state estimation models for real time battery management systems.

Parameter identification and co-estimation of state-of-charge of Li-ion battery in real-time on Internet-of-Things platform

An accurate estimate of the battery’s state-of-charge (SoC) is important for extending its service life and ensuring its protection. Based on a second-order equivalent circuit model, a joint estimator of online parameter identification and an improved adaptive extended Kalman filter (AEKF) for Li-ion battery SoC is developed, which tackles the problem of ill-conditioning while estimating the SoC in real-time. A three-state decoupled parameter estimation approach is designed to compute the best parameters:(i) ohmic resistance estimator with recursive least square (RLS) method, (ii) Tikhonov regularization has been introduced with the Kalman filter (KF) for the estimation of the fast time constant which improves the conditioning of the system and (iii) the measure of the slow time constant using lookup table-based method. Finally, an improved AEKF with fading weight factor mechanism is used to achieve closed-loop feedback of SOC estimation with parameter estimation, confirming the algorithm’s correctness and speed of convergence. The Chi-square test is incorporated into the AEKF to improve the rationale of the adaptive measurement covariance introduction timing. A variety of experimentally generated trending current patterns were used to validate the approach under various complex operating conditions. Furthermore, the proposed model-based diagnostic techniques have been integrated into a cloud-based system in this article, ensuring continuous and accurate battery state monitoring. All battery-related data is measured and transmitted to the cloud via the Internet of Things (IoT), where the proposed diagnostic algorithms analyze the data and estimate the SoC. As the proposed estimation approach features a low computational expense, it can be used in real-time industrial applications.

Design of peer-to-peer energy trading in transactive energy management for charge estimation of lithium-ion battery on hybrid electric vehicles

Transactive energy is a next big thing in the energy sector which considerably transform the association between utility, consumer, and the environment. Transactive energy management (TEM) system comprises a collection of economical and controlling strategies which helps to dynamically balance the power infrastructure. Peer-to-Peer (P2P) trading becomes familiar, which mainly based on the process of power generation and consumption is completely decentralized. At the same time, Hybrid electric vehicles (HEV) have become an important technology to accomplish energy efficiency and environmental sustainability. Battery Management System (BMS) evaluates the power and State of Charge (SOC), confirms the well-being depending upon the measurement. Accurate SOC estimation is crucial to ensure the unfailing functioning of Li-ion battery that is mainly employed in HEVs. A reliable SOC prediction model is needed to assure the precise measurement of the residual driving range of the vehicle and appropriate battery balancing. In this view, this paper presents an optimal machine learning based SOC estimation (OML-SOCE) model for HEVs in TEM. The OML-SOCE is aimed for estimating an accurate capacity of SOC of Li-ion batteries on HEVs. The OML-SOCE technique involves a two stage process namely stacked sparse autoencoder (SSAE) based prediction and salp swarm algorithm (SSA). At the first stage, SSAE is used for the prediction of SOC. Sparse auto-encoder (AE) is an enhanced AE technique that improves any some sparsity restrictions in the hidden layer of standard AE. A stack of multiple sparse AE forms a deep network framework that is named as SSAE. Next, in the second stage, the parameters involved in the SSAE are optimally adjusted by the use of SSA in such a way that the prediction performance can be considerably improved. A wide range of experiments take place and the results are investigated under varying temperature levels. The experimental outcomes showcased the supremacy of the presented technique the recent techniques with respect to different measures.

A novel combined estimation method of online full‐parameter identification and adaptive unscented particle filter for Li ‐ion batteries SOC based on fractional‐order modeling

Accurate estimation of the state of charge (SOC) of Li-ion battery can ensure the reliability of the storage system. A combined estimator of online full-parameter identification and adaptive unscented particle filter for Li-ion battery SOC based on an improved fractional-order model is proposed, which overcomes the shortcomings of the traditional SOC cumulative error and the difficulty of OCV acquisition. The proposed adaptive fractional unscented particle filter algorithm introduces fractional parameters as hidden parameters and reduces the complexity of the algorithm iteration by reducing the number of particles. At the same time, the noise adaptive algorithm based on the residual sequence can solve the divergence problem of the filter and improve the adaptability of the algorithm. To verify the feasibility of the algorithm under complex operating conditions, the urban dynamometer driving schedule dynamic working conditions of Li-ion batteries are verified. The experimental results show that the evaluation index of the algorithm is the best, the RMSE is 0.67%, and the SOC estimation is more accurate. It shows that the algorithm has strong robustness and fast convergence.

A novel adaptive dual extended Kalman filtering algorithm for the Li‐ion battery state of charge and state of health co‐estimation

Accurate prediction of the state of health (SOH) of Li-ion battery has an important role in the estimation of battery state of charge (SOC), which can not only improve the efficiency of battery usage but also ensure its safety performance. The battery capacity will decrease with the increase of charge and discharge times, while the internal resistance will become larger, which will affect battery management. The capacity attenuation characteristics of Li-ion batteries are analyzed by aging experiment. Based on the equivalent circuit model and online parameter identification, a novel adaptive dual extended Kalman filter algorithm is proposed to consider the influence of the battery SOH on the estimation of the battery SOC, and the SOC and SOH of the Li-ion battery are estimated collaboratively. The feasibility and accuracy of the model and algorithm are verified by experiments. The results show that the algorithm has good convergence and tracking. The maximum error in the estimation of the SOC is 2.03%, and the maximum error of the Ohmic resistance is 15.3%. It can better evaluate the SOH and SOC of Li-ion battery and reduce the dependence on experimental data, providing a reference for the efficient management of Li-ion batteries.

Estimation of State of Charge (SoC) Using Modified Coulomb Counting Method With Open Circuit Compensation For Battery Management System (BMS)

Abstract— Battery is one of the important components in the development of renewable energy technology. This paper presents a method for estimating the State of Charge (SoC) for a 4Ah Li-ion battery. State of Charge (SoC) is the status of the capacity in the battery in the form of a percentage which makes it easier to monitor the battery during use. Coulomb calculations are widely used, but this method still contains errors during integration. In this paper, SoC measurement using Open Circuit Voltage Compensation is used for the determination of the initial SoC, so that the initial SoC reading is more precise, because if the initial SoC reading only uses a voltage sensor, the initial SoC reading is less precise which affects the next n second SoC reading. In this paper, we present a battery management system design or commonly known as BMS (Battery Management System) which focuses on the monitoring function. BMS uses a voltage sensor in the form of a voltage divider circuit and an ACS 712 current sensor to send information about the battery condition to the microcontroller as the control center. Besides, BMS is equipped with a protection relay to protect the battery. The estimation results of the 12volt 4Ah Li-ion battery SoC with the actual reading show an error of less than 1%.Keywords—Battery Management System, Modified Coulomb Counting, State of Charge.