Accurate State Of Charge Estimation Research Articles

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 with Spatial Attention-Based CONV-LSTM for SOC Estimation of Lithium-Ion Batteries

Accurate estimation of the state of charge (SOC) is an indispensable part of a vehicle management system. The accurate estimation of SOC can ensure the safe and reliable operation of the vehicle management system. With the development of intelligent transportation systems (ITS), vehicles can not only obtain the dynamic changes inside the battery through sensors, but also obtain the traffic information around the vehicle through vehicle–road collaboration. In addition, the development of onboard graphic processing units (GPUs) and Internet of Vehicles (IOV) technology make the computing power of vehicles no longer limited by hardware, which makes neural networks applied to the intelligent control of vehicles. Aiming at the problem that the traditional network cannot effectively obtain the complex spatial information of sample attributes, we developed an attention-based CONV-LSTM module for SOC prediction based on a convolutional neural network (CNN) and a long short-term memory (LSTM) network. Different from the traditional LSTM network, the algorithm not only considers the temporal correlation of the data stream, but also captures the spatial correlation information of the input data by convolution. It then uses the different weights, automatically assigned by the attention mechanism, to correctly distinguish the importance of different input data streams. In order to verify the validity of the model, this paper selects the degradation data set of the aeroengine as the verification data set. Experiments show that the proposed model has achieved good results. Finally, the proposed model is applied to the actual vehicle running data, and the effectiveness of the proposed model is verified by comparing it with the Multi-Layer Perceptron (MLP), LSTM, and CNN-LSTM models.

State of charge estimation of lithium-ion battery based on multi-input extreme learning machine using online model parameter identification

The state of charge (SOC) is important for ensuring both battery safety and life. Furthermore, the SOC is required to estimate other battery states such as state of power (SOP) and state of health (SOH). Because the SOC is determined by estimation rather than observation, it is important to establish a proper estimation method. In this paper, an equivalent circuit model (ECM) was first constructed through online parameter extraction. Online parameter identification was based on a recursive least squares (RLS) method to input the various internal information regarding the battery into the extreme learning machine to achieve accurate SOC estimation. Second, to deliver a highly accurate SOC estimation of lithium-ion batteries (LiBs), the multi-input extreme learning machine (MI-ELM) method based on an online model parameter identification technique was applied to the SOC estimation of LiBs. Finally, experiments were conducted under various operating conditions to assess the performance of the proposed method. Compared with other estimation methods such as the extended Kalman filter (EKF), the ordinary extreme learning machine (ELM), the adaptive square root extended Kalman filter (ASREKF), the autoencoder neural network with long short-term memory neural network (AUTOENCOD-LSTM), the artificial neural network and unscented Kalman filter (NN&UKF), and the gravitational search algorithm-based ELM (ELM-GSA) in terms of the mean absolute error (MAE) and the root mean square error (RMSE), the proposed MI-ELM method achieved 60.00 %, 88.83 %, 52.50 %, 80.00 %, 84.55 %, and 79.64 % of the maximum performance improvement, respectively.

Dual fuzzy-based adaptive extended Kalman filter for state of charge estimation of liquid metal battery

Liquid metal batteries (LMBs) are alternatives to conventional lithium-ion batteries due to their specific benefits including high current density and long cycle life. Accurate state of charge (SOC) estimation is an important evaluation index for the battery management system (BMS), which is of great significance to ensure the safe operation of batteries. However, the estimation accuracy of SOC is influenced by many factors, including self-aging and external operating environment changes. Therefore, an online battery model with real-time parameter updates is necessary for accurate SOC estimation. In this paper, a novel dual fuzzy-based adaptive extended Kalman filter (DFAEKF) method is proposed for the SOC estimation of LMBs. Firstly, a second-order RC equivalent circuit model is established to describe the battery's behavior. The forgetting factor recursive least squares (FFRLS) is utilized to identify the model parameters and reconstruct the battery open circuit voltage (OCV). Secondly, the dual adaptive extended Kalman filter (DAEKF) is derived from the battery model. And an intelligent noise estimator is designed based on a fuzzy inference system, which adaptively adjusts the length of the residual innovation sequence (RIS), to update the noise covariance. Finally, the DFAEKF algorithm is proposed for the battery SOC and parameter co-estimation. The online estimated ohmic resistance is employed to calculate the state of health (SOH) of the battery. The proposed DFAEKF is verified through different experiments and compared to conventional algorithms. Experimental results show that the DFAEKF has higher accuracy (error < 1 %) and stronger robustness. The proposed method can also be applied to other model-based state estimation areas.

Improved unscented Kalman filter for state of charge estimation of lithium-ion battery with one-step randomly measurement loss and inaccurate noise covariance matrices

Accurate state of charge estimation is of great significance for optimizing the management of lithium batteries. Due to the influence of actual environmental factors, random measurement loss and inaccurate noise covariance matrices will appear in the estimation process. Especially when the two situations occur at the same time, the estimation of SOC will be seriously affected. Thus, an improved unscented Kalman filter is presented to solve the above problems in this paper. First, the state space model of nonlinear system is constructed, which can fully reflect one-step random measurement loss, and the nominal noise covariance matrix is used to reflect the uncertainty. Second, as the prior probability density function of the parameters, the Beta prior and inverse Wishart prior are used in the derivation of the joint probability density function. Finally, these parameters are derived based on variational Bayesian theory, the updated state vector is obtained. Simulation results shows that the proposed method can deal with random measurement loss and inaccurate noise covariance matrices and the high accuracy in dynamic stress test and federal urban driving schedule.

Battery state of charge estimation using temporal convolutional network based on electric vehicles operating data

Accurate estimation of state of charge (SOC) is crucial for battery management system (BMS). Since most of the existing estimation methods are based on laboratory data, the accuracy of battery SOC estimation in real driving scenarios is difficult to be guaranteed. This paper proposes a method to estimate battery SOC using temporal convolutional network based on real-world operating data of electric vehicles (EVs). In view of the characteristics of large volume and high dimensionality of real-world EVs data, attention mechanisms are added to the time steps and input feature dimensions of the data, respectively, to further improve the model estimation accuracy. Meanwhile, this paper proposes a new method to correct the SOC value provided by the cloud platform through developing an improved capacity decay model applicable to the real-world EVs data. Then the corrected SOC is used as the label for neural network training. Experiments are conducted using one-year EVs data as the testing dataset, and the results show that the proposed model performances well, with root mean square error (RMSE) <0.67 % in spring and 0.99 % in winter. In comparison with other neural network models commonly used to predict battery SOC, the results show that the proposed model predicts more stable and accurate results.

Co-Estimation of State-of-Charge and State-of-Health for Lithium-Ion Batteries Considering Temperature and Ageing

State-of-charge (SOC) estimation of lithium-ion batteries (LIBs) is the basis of other state estimations. However, its accuracy can be affected by many factors, such as temperature and ageing. To handle this bottleneck issue, we here propose a joint SOC-SOH estimation method considering the influence of the temperature. It combines the Forgetting Factor Recursive Least Squares (FFRLS) algorithm, Total Least Squares (TLS) algorithm, and Unscented Kalman Filter (UKF) algorithm. First, the FFRLS algorithm is used to identify and update the parameters of the equivalent circuit model in real time under different battery ageing degrees. Then, the TLS algorithm is used to estimate the battery SOH to improve the prior estimation accuracy of SOC. Next, the SOC is calculated by the UKF algorithm, and finally, a more accurate SOH can be obtained according to the UKF-based SOC trajectory. The battery-in-the-loop experiments are utilized to verify the proposed algorithm. For the cases of temperature change up to 35 °C and capacity decay up to 10%, our joint estimator can achieve ultra-low errors, bounded by 2%, respectively, for SOH and SOC. The proposed method paves the way for the advancement of battery use in applications, such as electric vehicles and microgrid applications.

An optimized long short-term memory-weighted fading extended Kalman filtering model with wide temperature adaptation for the state of charge estimation of lithium-ion batteries

Accurate state of charge (SOC) estimation at different operating temperatures is essential for the reliable and safe operation of battery management systems (BMS) for lithium-ion batteries in electric vehicles (EVs). In this paper, an optimized long-short-term memory-weighted fading extended Kalman filtering (LSTM-WFEKF) model with wide temperature adaptation is proposed as a temperature-conditioned model for SOC estimation. Firstly, the input datasets are categorized based on the operating temperatures for EVs in the United States Advanced Battery Consortium manual: cold (−10 °C), normal (25 °C), and hot (50 °C) temperatures and optimized with an attention mechanism for faster training of the LSTM model to cross-train and test to specifically study the effects of temperature on the SOC estimation through a transfer learning mechanism. Secondly, the SOC estimated by the LSTM model is input into a WFEKF method, which introduces adaptive weighing and fading factors to correct, denoise, and optimize the final SOC for each temperature variation under complex working conditions. Finally, the results show that the training and testing temperatures have distinctive SOC effects using the LSTM model. Also, the proposed LSTM-WFEKF model estimates the SOC with overall best mean absolute error (MAE), root mean square error (RMSE), and R-squared (R2) values of 0.0697%, 0.0784%, and 99.9965%, respectively, under different temperatures and complex working conditions, which is optimal compared to other existing models. Based on the MAE, RMSE, and R2 values under different operating temperatures and complex working conditions, this paper concludes that the 25 °C training dataset ensures a more accurate SOC estimation. Meanwhile, the −10 °C and 50 °C training datasets cause more and less noisy estimates, respectively. The proposed LSTM-WFEKF model has wide temperature and working condition adaptability for real-time BMS applications in EVs.

Bayesian information criterion based data-driven state of charge estimation for lithium-ion battery

Accurate state of charge (SOC) estimation is essential for the safe and reliable operation of Li-ion batteries. To solve the problem of poor generalisation caused by over-fitting, this paper presents a combination algorithm based on feature selection to estimate battery SOC. Firstly, a portion of the features is extracted from the extended Kalman filtering (EKF) results. It forms the set of features to be selected with four other measured features. Secondly, the optimal feature subset is adopted by designing a wrapped feature screening framework based on the Bayesian information criterion (BIC). Finally, the selected combination of features is adopted to train the support vector regression (SVR) model, which is applied to the battery SOC estimation. The experimental results reveal that the combination strategy of EKF and SVR improves the accuracy of SOC estimation. The optimal SVR model based on the feature selection criterion shows better generalisation. Better estimation results in four driving conditions are achieved, and the root-mean-square error of the battery SOC estimation is decreased by at least 64.1 % and 56.5 % compared to the EKF algorithm and SVR algorithm driven by full feature, respectively.

A novel method for determination of SOC of Lead-acid battery in E-Rickshaw application

Electric rickshaws are a popular mode of short-distance transportation in cities and an alternative to petrol- or diesel-powered auto-rickshaws because of their low capital and running cost and less human effort compared to pedal rickshaws. E Rickshaws can operate if there is energy or charge in the battery. State of Charge (SoC) indicates the amount of energy left in the batter. Accurate estimation of SoC is essential for estimating how long the vehicle can travel apart from overcharge and under-discharge protection resulting in life enhancement. Conventional methods of SoC prediction are based on the estimation of open-circuit voltage. These are inaccurate and time-consuming and become erroneous as the battery ages. A method for quick determination of SoC using a high current short-time discharge is proposed for both old and new batteries. Experimental investigations on 90Ah C10 batteries using the said methodology can determine the correct state of charge in any condition with a maximum error of ± 8 %.

Voltage-stress-based state of charge estimation of pouch lithium-ion batteries using a long short-term memory network

Accurate state-of-charge (SOC) estimation is critical to the energy management of lithium-ion batteries, which are increasingly employed in electric vehicles (EVs). In this work, a novel SOC estimation method of a battery based on voltage and stress measurements is proposed with the utilization of machine learning techniques. To evaluate the necessity of stress measurements for SOC estimation, a series of stress characteristic experiments are designed to probe the relationship between mechanical responses and the SOC. On this basis, the stress is taken as an auxiliary for the SOC estimation. Then, a typical long short-term memory (LSTM) network is introduced to map battery measurement signals directly to the battery SOC, where the voltage, stress, and current are selected as input vectors. The proposed method is evaluated under constant current (CC), dynamic stress test (DST), and urban dynamometer driving schedule (UDDS) conditions. Experimental results show that the maximum absolute error (MAE) and root mean square error (RMSE) can be bounded within 0.34 % and 0.45 %, which demonstrates that the voltage-stress-based method can achieve good SOC estimation accuracy. Therefore, the proposed method provides new insight into the SOC estimation of pouch lithium-ion batteries.

SOC estimation based on the gas‐liquid dynamics model using particle filter algorithm

Lithium-ion battery state of charge (SOC) estimation is the basis for other state estimations in electric vehicle (EV) battery management systems. Accurate SOC estimation can prevent overcharge and discharge of the battery, improve discharge efficiency and prolong cycle life. A high-precision SOC estimation method based on the gas-liquid dynamics (GLD) model is presented. The standard particle filter (PF) and the unscented particle filter (UPF) algorithm are successively used for optimizing and improving the chattering problem of the original GLD model. The results show that the two kinds of PF algorithms can effectively solve the chattering problem of the GLD model and improve the estimation accuracy. Among them, UPF has higher accuracy and the maximum estimation error is less than 2%. The PF method for estimating SOC introduced in this paper has the characteristics of high precision and large amount of calculation. Due to the above characteristics, this method can be used as a cloud SOC calculation method to periodically correct the local SOC of the vehicle.

State of charge estimation for lithium-ion batteries based on fractional order multiscale algorithm

The accuracy of state of charge (SoC) plays a critical role in lithium-ion batteries management system, whereas it gradually reduces as ohmic internal resistance growing and maximum available capacity decreasing with battery aging, which is still an open and challenging problem in SoC estimation. To tackle this issue, a multi-scale estimation algorithm is proposed by incorporating fractional order adaptive extended Kalman filter and variable forgetting factor recursive least square (FOAEKF-VFFRLS). Firstly, in order to describe the physicochemical reactions inside battery accurately, a fractional order model of lithium-ion battery with its discrete form is introduced, and the resistor-capacitor (RC) values as well as the order of capacitors are obtained by employing ant colony algorithm (ACA). Secondly, for enhancing SoC estimation accuracy, a fractional order adaptive extended Kalman filter algorithm (FOAEKF) is proposed to continuously modify the noise statistical characteristics while utilizing observation data information. Thirdly, to predict ohmic internal resistance and maximum available capacity for battery, a variable forgetting factor recursive least square algorithm (VFFRLS) is established based on the fact that forgetting factor determined by individual data error would be suddenly changes in the case of inaccurate or perturbed data. Besides, to further improve the accuracy of SoC estimation, a multi-scale estimation approach is proposed in light of the slow variation of battery internal resistance and the fast change of SoC. Finally, through comparative experiments, it is proved that the proposed approach has comprehensive strengths. First, the minimum root mean square error (RMSE) of SoC estimation error is controlled within 0.52 %. Second, terminal voltage error is kept small while limiting SoC errors within narrow ranges. Third, favorable robustness in terms of SoC estimation is achieved despite SoC initial value is mis-initialized.

A Hybrid Method for State-of-Charge Estimation for Lithium-Ion Batteries Using a Long Short-Term Memory Network Combined with Attention and a Kalman Filter

A battery management system (BMS) is an important link between on-board power battery and electric vehicles. The BMS is used to collect, process, and store important information during the operation of a battery pack in real time. Due to the wide application of lithium-ion batteries in electric vehicles, the correct estimation of the state of charge (SOC) of lithium-ion batteries (LIBS) is of great importance in the battery management system. The SOC is used to reflect the remaining capacity of the battery, which is directly related to the efficiency of the power output and management of energy. In this paper, a new long short-term memory network with attention mechanism combined with Kalman filter is proposed to estimate the SOC of the Li-ion battery in the BMS. Several different dynamic driving plans are used for training and testing under different temperatures and initial errors, and the results show that the method is highly reliable for estimating the SOC of the Li-ion battery. The average root mean square error (RMSE) reaches 0.01492 for the US06 condition, 0.01205 for the federal urban driving scheme (FUDS) condition, and 0.00806 for the dynamic stress test (DST) condition. It is demonstrated that the proposed method is more reliable and robust, in terms of SOC estimation accuracy, compared with the traditional long short-term memory (LSTM) neural network, LSTM combined with attention mechanism, or LSTM combined with the Kalman filtering method.

State of charge estimation of lithium‐ion battery based on extended Kalman filter and unscented Kalman filter techniques

AbstractLithium‐ion batteries (LIBs) are currently the most widely used energy storage technologies in electric vehicles (EVs) due to their higher power density, greater energy density, longer life, fast charging capabilities, and low discharge rate. The brain of rechargeable batteries is referred to be the battery management system (BMS). The operation of the battery is monitored &amp; regulated by an electronic system called BMS. Most significantly, it prevents the battery from exceeding its safety limits. One of the important parameters of BMS is the state of charge (SOC); it determines the battery's remaining capacity of accessible stored energy, battery lifespan, and cell balancing. As a result, an accurate SOC calculation is essential for BMS, and it is used for safer operations that avoid overcharging and discharging for the optimum battery life. The primary ambition of this article is to find an accurate and robust SOC estimation based on extended Kalman filter (EKF) and unscented Kalman filter (UKF) techniques. RMS error, average error, minimum error, and maximum error are used to compare the EKF method and UKF method. Two resistor‐capacitor (2RC) equivalent circuit model has been considered for the dynamic performance of the battery. The waveforms are displayed, and errors are simulated for SOC estimate using the EKF and UKF algorithms, taking the same initial SOC into consideration at a time. The UKF algorithm outperforms the EKF approach in terms of dynamic results when comparing the two methods.

State of charge estimation of lithium-ion battery based on improved forgetting factor recursive least squares-extended Kalman filter joint algorithm

In order to solve the problem that forgetting factor recursive least squares (FFRLS) is prone to abnormal jitter and even divergence under complex working conditions, improved forgetting factor recursive least squares based on dynamic constraint and parameter backtracking is proposed. A joint algorithm of improved forgetting factor recursive least squares and extended Kalman filter (EKF) is used to estimate the state of charge (SOC) of lithium-ion battery. Firstly, parameters of Thevenin equivalent circuit model are identified on-line by the improved FFRLS considering dynamic constraint and parameter backtracking, and then the SOC of lithium-ion battery is estimated by extended Kalman filter. The results show that the improved forgetting factor recursive least squares has high accuracy of battery model parameters identification and the joint algorithm also has high accuracy and robustness of SOC estimation under dynamic stress test (DST) condition, the maximum absolute SOC estimation error is 2.49 % and the average absolute SOC estimation error is 1.39 %.

State-of-charge estimation for lithium-ion batteries of electric vehicle based on sensor random error compensation

Accurate state of charge (SOC) estimation is essential for ensuring the security and efficiency of lithium-ion batteries. Generally, the signal detected by a sensor inevitably has noise and drift. Therefore, the prediction process will indefinitely accumulate this random error, resulting in lower estimation accuracy, which is particularly prominent when the performance of vehicle sensors is poor. To solve this problem, a method combining an adaptive extended Kalman filter (AEKF) with a long short-term memory (LSTM) network is proposed in this paper. The original data with random errors observed in each period is denoised by the AEKF; then, the processed values are served as the input of the LSTM neural network. In this way, it avoids the disadvantage that the LSTM depends too much on the accuracy of training data and retains the advantages of the high robustness of the AEKF and strong nonlinear characteristics of the LSTM network. The proposed method is compared with an LSTM method without pre-processing the input data on the New European Driving Cycle (NEDC) and the latest World Light Vehicle Test Cycle (WLTC) datasets. Subsequently, the experimental results are analysed and compared with other algorithms under various initial noise, temperatures, and unknown initial SOC conditions. The results show that the accuracy of SOC estimation is improved a lot in the case of compensating the random input error by the AEKF in advance in comparison with feeding it into a neural network algorithm directly, converging quickly under inaccurate initial conditions and having strong robustness against noise and other factors. In particular, the root-mean-square error is less than 0.6 %, and the maximum error is less than 1.6 %.

A Novel Bias Compensated Recursive Least Squares and Multi Innovation Unscented Kalman Filtering Algorithm Method for Accurate State of Charge Estimation of Lithium-ion Batteries

A Novel Bias Compensated Recursive Least Squares and Multi Innovation Unscented Kalman Filtering Algorithm Method for Accurate State of Charge Estimation of Lithium-ion Batteries

A Review of Critical State Joint Estimation Methods of Lithium-Ion Batteries in Electric Vehicles

Battery state of charge (SOC), state of health (SOH), and state of power (SOP) are decisive factors that influence the energy-management system (EMS) performance of electric vehicles. However, the accurate estimation of SOC, SOH, and SOP remains a challenge due to the high nonlinearity of the battery dynamic characteristics and the strong coupling among the states. In this paper, different methods of single-state and two-state joint estimation are classified and discussed, including SOC/SOH and SOC/SOP joint estimation methods, and their advantages and limitations are analyzed. On this basis, key issues of joint multi-state estimation are discussed, and suggestions for future work are made.

State-of-Charge Estimation for Lithium-Ion Batteries Using Residual Convolutional Neural Networks.

State-of-charge (SOC) is a relative quantity that describes the ratio of the remaining capacity to the present maximum available capacity. Accurate SOC estimation is essential for a battery-management system. In addition to informing the user of the expected usage until the next recharge, it is crucial for improving the utilization efficiency and service life of the battery. This study focuses on applying deep-learning techniques, and specifically convolutional residual networks, to estimate the SOC of lithium-ion batteries. By stacking the values of multiple measurable variables taken at many time instants as the model inputs, the process information for the voltage or current generation, and their interrelations, can be effectively extracted using the proposed convolutional residual blocks, and can simultaneously be exploited to regress for accurate SOCs. The performance of the proposed network model was evaluated using the data obtained from a lithium-ion battery (Panasonic NCR18650PF) under nine different driving schedules at five ambient temperatures. The experimental results demonstrated an average mean absolute error of 1.260%, and an average root-mean-square error of 0.998%. The number of floating-point operations required to complete one SOC estimation was 2.24 × 106. These results indicate the efficacy and performance of the proposed approach.