Battery State Estimation Research Articles

Hybrid Modeling of Lithium-Ion Battery: Physics-Informed Neural Network for Battery State Estimation

Accurate forecasting of the lifetime and degradation mechanisms of lithium-ion batteries is crucial for their optimization, management, and safety while preventing latent failures. However, the typical state estimations are challenging due to complex and dynamic cell parameters and wide variations in usage conditions. Physics-based models need a tradeoff between accuracy and complexity due to vast parameter requirements, while machine-learning models require large training datasets and may fail when generalized to unseen scenarios. To address this issue, this paper aims to integrate the physics-based battery model and the machine learning model to leverage their respective strengths. This is achieved by applying the deep learning framework called physics-informed neural networks (PINN) to electrochemical battery modeling. The state of charge and state of health of lithium-ion cells are predicted by integrating the partial differential equation of Fick’s law of diffusion from a single particle model into the neural network training process. The results indicate that PINN can estimate the state of charge with a root mean square error in the range of 0.014% to 0.2%, while the state of health has a range of 1.1% to 2.3%, even with limited training data. Compared to conventional approaches, PINN is less complex while still incorporating the laws of physics into the training process, resulting in adequate predictions, even for unseen situations.

Deep learning to estimate lithium-ion battery state of health without additional degradation experiments

State of health is a critical state which evaluates the degradation level of batteries. However, it cannot be measured directly but requires estimation. While accurate state of health estimation has progressed markedly, the time- and resource-consuming degradation experiments to generate target battery labels hinder the development of state of health estimation methods. In this article, we design a deep-learning framework to enable the estimation of battery state of health in the absence of target battery labels. This framework integrates a swarm of deep neural networks equipped with domain adaptation to produce accurate estimation. We employ 65 commercial batteries from 5 different manufacturers to generate 71,588 samples for cross-validation. The validation results indicate that the proposed framework can ensure absolute errors of less than 3% for 89.4% of samples (less than 5% for 98.9% of samples), with a maximum absolute error of less than 8.87% in the absence of target labels. This work emphasizes the power of deep learning in precluding degradation experiments and highlights the promise of rapid development of battery management algorithms for new-generation batteries using only previous experimental data.

Joint Estimation of State of Charge and State of Health of Lithium Ion Battery

Abstract Overcharge and discharge of power battery not only increase the battery loss but also lead to fire and other accidents under harsh environmental conditions. Accurate estimation of battery parameters and status is an important reference in the battery management system to prevent battery overcharge and discharge. In this article, the following studies are carried out by focusing on the time separation scale and estimating parameters and state values online based on the improved particle filter: (1) The unscented transform and multi-innovation were applied to the particle filter to optimize the particle distribution and update the status value from the historical information, and the multi-innovation unscented particle filter was formed to estimate the state of battery charge. (2) Considering the influence of parameter variation on the estimation of battery state of charge (SOC). Due to the slow change characteristics of parameters and fast change characteristics of states, the parameters and states are jointly estimated from macro and micro time scales, respectively. The capacity change estimated by the unscented particle filter is used to characterize the battery health state, and finally, the joint estimation of battery SOC and state of health (SOH) is formed; (3) Three different working conditions are used to verify the algorithm. The joint algorithm accurately estimates the real-time changes of SOC and SOH, and the average error of SOC is less than 0.5%, which confirms the high accuracy of the joint algorithm.

A Two-Stage Optimization Framework for Fast Lithium-Ion Battery Impedance Measurement

Online monitoring of battery impedance is of increasing interest as it brings rich information for battery state estimation and fault diagnosis, however, the existing techniques are limited by high implementation costs and poor accuracy. This letter addresses the challenges by proposing a novel two-stage optimization framework to form a binary sequence with the optimal power proportion in specified harmonics for fast battery impedance measurements. The Optimized Binary Sequence (OBS) with finite amplitude levels can greatly simplify the implementation, and also the optimized power spectrum has superb accuracy to obtain the frequency responses. Experimental studies confirm the superiority of the proposed method in comparison with three existing techniques under various circumstances.

Remaining driving range prediction for electric vehicles: Key challenges and outlook

AbstractRemaining driving range (RDR) research has continued to consistently evolve with the development of electric vehicles (EVs). Accurate RDR prediction is a promising approach to alleviate distance anxiety when power battery technology is not yet fully matured. This paper first introduces the research motivation of RDR prediction, summarizes the previous research progress, and classifies the influencing factors of RDR. Second, conduct research and analysis on the physical model of EVs, mainly including battery and vehicle models. Based on the physical model, the energy flow problem of EVs is analyzed and discussed. Third, four key challenges of RDR prediction are summarized: battery state estimation, driving behavior classification and recognition, driving condition prediction and speed prediction, and RDR calculation method. Finally, given the four challenges faced by RDR, a driving range prediction method based on vehicle‐cloud collaboration is proposed, which combines the advantages of cloud computing and machine learning to provide further research trends.

적응형 확장 칼만 필터를 활용한 무인기의 배터리 충전 상태 추정

본 연구에서는 적응형 확장 칼만 필터를 활용하여 무인기의 배터리 충전 상태를 추정하는 시스템을 제안한다. 배터리 전압과 충전 상태의 관계를 표현할 수 있는 Shepherd 모델을 적용한다. Shepherd 모델을 구성하는 파라미터는 시험 비행데이터를 활용하여 비선형 최적화 방법으로 식별한 후 그 결과를 적용한다. 제안한 시스템을 검증하기 위해 Panasonic 18650PF 공개 데이터와 비행 데이터를 활용한다. 성능 지표는 측정된 충전 상태와 추정 충전 상태의 오차로 설정하고, 확장 칼만 필터와 제안한 필터의 오차를 비교하여 평가한다. 그리고 무인기의 최대 비행시간을 확인하기 위한 목적으로 수행된 3 소티의 비행 데이터를 제안한 시스템 검증에 활용한다. Panasonic 18650PF 공개 데이터와 비행 데이터 모두에서 확장 칼만 필터의 성능과 비교하여 제안한 시스템의 충전 상태 추정 정확도가 향상됨을 확인할 수 있다.

State of Charge and State of Health Estimation of Lithium-Ion Battery Packs With Inconsistent Internal Parameters Using Dual Extended Kalman Filter

Abstract The internal battery parameters of the lithium-ion battery energy storage system may be inconsistent due to different aging degrees during the operation, and the thermal effect can also threaten the safety of the system. In this paper, based on the second-order resistor–capacitor equivalent circuit model and the dual extended Kalman filter (DEKF) algorithm, an electrical simulation model of a LIB pack with inconsistent parameters considering the thermal effect is established, in which state of charge (SOC) and state of health (SOH) are estimated using DEKF, while the temperature is calculated by a thermal module. The simulation results show that the DEKF algorithm has a good effect on battery state and parameter estimation, with the root-mean-square error of voltage is lower than 0.01 V and SOC mean absolute error (MAE) is below 1.50%, while SOH error is 3.37%. In addition, the thermal module can provide an accurate estimation of the inconsistent temperature rise of the battery pack, and the MAE between the model-calculated temperature and the experiment is no more than 6.60%. The results provide the basic data for the scale-up of the electrothermal co-simulation model of the LIB energy storage system.

Co-estimation of state of charge and internal temperature of pouch lithium battery based on multi-parameter time-varying electrothermal coupling model

Accurate estimation of battery state of charge (SOC) and internal temperature can improve battery performance and safety. In this process, the accuracy of the battery's electrical model and thermal model and the applicability and robustness of the estimation algorithm are key. In this paper, a multi-parameter time-varying electrothermal coupling model for pouch batteries is established. The model takes into account the coupling relationship between the battery SOC and temperature changes, as well as the heat exchange between multiple surfaces of the battery and the external environment. And then through the improved hybrid pulse power characterization (HPPC) test to realize the identification of model parameters, and the changing external heat transfer conditions could be identified online. At last, an adaptive strong tracking square root cubature Kalman filter (AST-SRCKF) algorithm was designed to estimate SOC and internal temperature simultaneously. The results were verified by three dynamic conditions. The conclusion is that the AST-SRCKF has high robustness and can be quickly corrected when the initial values of SOC and temperature are unknown. And the estimation accuracy is higher than extended Kalman filter (EKF) and dual Kalman filter (DEKF). Furthermore, the simulation efficiency is improved by more than 6 % compared with EKF, and by more than 11 % compared with DEKF, which is more conducive to online real vehicle applications.

Co-estimation of lithium-ion battery state-of-charge and state-of-health based on fractional-order model

Accurate estimation of lithium-ion battery state of charge and state of health has been the focus of research in battery management systems. Effectively improving the estimation accuracy of battery state parameters is crucial for the safe and stable driving of electric vehicles. In this paper, based on the fractional-order model, a multi-scale cooperative estimation method based on the dual adaptive Unscented Kalman filter is proposed. Firstly, according to the fast and slow time-varying characteristics of the measured parameters, the dual filters perform state estimation of SOC and SOH according to different time scales, alternately updating the information parameters by nesting them with each other and optimizing the noise terms with an adaptive algorithm. Meanwhile, the influence of temperature on the battery parameters is considered, and experimental analysis of the battery is carried out using a variety of different temperatures and operating conditions, as well as the accuracy and stability of the algorithm are verified by simulation. The experimental and simulation results show that the co-estimation method has good accuracy and robustness in both high and low-temperature environments.

On-Line Multi-Time Scale Adaptive Parameter Identification Based on Improved Lithium-Ion Batteries Hysteresis Characteristic-Electrical Equivalent Circuit Modeling

Accurate identification of model parameters is a key aspect of lithium battery state estimation. To accurately identify battery model parameters, this paper establishes Hysteresis Characteristic-Electrical Equivalent Circuit (HC-EEC) modeling by analyzing the influence of the hysteresis effect on the battery State of Charge (SOC). For the high-precision identification of battery model parameters, an Online Multi-Time Scale Adaptive Parameter Identification Strategy (OM-TSAPIS) is proposed in this paper. According to the different dynamic response links in the HC-EEC model, the strategy performs parameter identification through different time scale links and uses the adaptive step size as the starting identification condition for the multi-time scale links, thereby improving the parameter identification accuracy of the HC-EEC model. The absolute average error of OM-TSAPIS was 0.0437 mV and 0.298 mV under the Urban Dynamometer Driving Schedule (UDDS) and Beijing Bus Dynamic Street Test (BBDST) conditions, respectively. Simulation results show that the identification accuracy of the proposed algorithm is high.

Electric vehicle battery state of health estimation using Incremental Capacity Analysis

The state of health (SOH) is an essential indicator for electric vehicle (EV) batteries. It is important to ensure a proper and safe operation of the battery and of great interest for the increasing second-hand market. As the battery is the most expensive EV component the value of the entire vehicle depends on the SOH. The purpose of this work is to develop and verify a non-invasive method to determine the SOH and study the capability of the incremental capacity analysis (ICA) method to estimate SOH of real EVs. The ICA technique is a common method for battery state estimation through the analysis of the voltage changes produced by the electrochemical reactions that occur during charging and discharging. In this work aging tests have been carried out on cells to determine the ICA features with the best performance to estimate the SOH. LMO/NMC-based batteries used in the BMW i3 have been aged and analyzed using ICA. The position of some of the features extracted from the incremental capacity (IC) curve was used to estimate the SOH of the batteries. Subsequently, in order to verify the method, two vehicles with different mileage were tested using similar conditions, obtaining an accurate SOH estimation with an overall root mean square error (RMSE) of 2%. The proposed method has shown to be effective both in cells and in commercial available vehicles, allowing the estimation of SOH during charging and without the need to interfere in the process.

Research on Multi-Time Scale SOP Estimation of Lithium–Ion Battery Based on H∞ Filter

Battery state of power (SOP) estimation is an important parameter index for electric vehicles to improve battery utilization efficiency and maximize battery safety. Most of the current studies on the SOP estimation of lithium–ion batteries consider only a single constraint and rarely pay attention to the estimation of battery state on different time scales, which can reduce the accuracy of SOP estimation and even cause safety problems. In view of this, this paper proposes a multi-time scale and multi-constraint SOP estimation method for lithium–ion batteries based on H∞ filtering. Firstly, a second-order RC equivalent circuit model is established with a ternary lithium–ion monolithic battery as the research object, and parameter identification is performed by using the recursive least squares method with a forgetting factor. Secondly, the H∞ filtering algorithm is applied to estimate the state of charge (SOC), and then the joint multi-time scale multi-constrained SOC-SOP estimation is performed. Finally, the joint estimation algorithm is validated under UDDS conditions. The mean absolute value relative error (MARE) of SOC estimation is 1.17%, and the MARE of SOP estimation at different time scales is less than 1.6%. The results indicate the high accuracy and strong robustness of the joint estimation method.

A state of health estimation framework based on real-world electric vehicles operating data

The accurate estimation of battery state of health (SOH) is essential to alleviate mileage anxiety for the drivers of electric vehicles (EVs) and achieve echelon use of the battery. The data collected in real-world reflects complex operating conditions, which make the study of real-world EVs possible. In this paper, a framework for the SOH estimation of real-world EVs considering parameters optimization is proposed. First, the battery's capacity is calculated utilizing the deep charging process, and the non-linear trend of capacity decay is described by Generalized Additive Models (GAM). Then, the internal and external health features characterizing battery aging are extracted from the EVs data. The quality of the health features is evaluated by correlation analysis, and the parameters of the health features are selected with Bayesian optimization. Finally, the PSO-ELM model for SOH estimation is established by optimizing the weights and biases of extreme learning machine (ELM) by particle swarm optimization (PSO). The experimental results show that the framework proposed is able to accurately estimate the SOH of the real-world vehicle battery pack with a maximum error of less than 0.4 %.

Review of State Estimation and Remaining Useful Life Prediction Methods for Lithium–Ion Batteries

The accurate estimation of the state of charge, the state of health and the prediction of remaining useful life of lithium–ion batteries is an important component of battery management. It is of great significance to prolong battery life and ensure the reliability of the battery system. Many researchers have completed a large amount of work on battery state evaluation and RUL prediction methods and proposed a variety of methods. This paper first introduces the definition of the SOC, the SOH and the existing estimation methods. Then, the definition of RUL is introduced, and the main methods are classified and compared. Finally, the challenges of lithium–ion battery state estimation and RUL prediction are summarized, and the direction for future development is proposed.

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.

Lithium-Ion Battery State of Charge Estimation Using Deep Neural Network

In an electric vehicle (EV), the battery management system (BMS) is crucial for managing the health and safety of the battery. The accurate estimation of battery state of charge (SOC) offers critical information about the battery’s remaining capacity. The SOC of the battery mainly depends on its non-linear internal parameters, battery chemistry, ambient temperature, aging factor etc. So, accurate SOC estimation is still a significant challenge. Many researchers have developed several model-based methods that are more complex to develop. Another approach is a data-driven based SOC estimation algorithm, which is less complex but requires more data and it may be inaccurate. In this context, this paper presents a robust and accurate SOC estimation algorithm for a Lithium-ion battery using a deep learning feed-forward neural network (DLFFNN) approach. The proposed algorithm accurately characterizes the battery’s non-linear behavior. To develop a robust SOC estimation algorithm, data is collected at different temperatures with 5% error in data (4 mV-voltage, 110 mA-current, 5∘C temperature) is added to battery datasets. The obtained results demonstrated that the performance of the proposed DLFNN is robust and accurate on different drive cycles with 1.14% Root mean squared error (RMSE), 0.66% mean absolute error (MAE), and 6.65% maximum error (MAX).

A Survey on Energy Storage: Techniques and Challenges

Intermittent renewable energy is becoming increasingly popular, as storing stationary and mobile energy remains a critical focus of attention. Although electricity cannot be stored on any scale, it can be converted to other kinds of energies that can be stored and then reconverted to electricity on demand. Such energy storage systems can be based on batteries, supercapacitors, flywheels, thermal modules, compressed air, and hydro storage. This survey article explores several aspects of energy storage. First, we define the primary difficulties and goals associated with energy storage. Second, we discuss several strategies employed for energy storage and the criteria used to identify the most appropriate technology. In addition, we address the current issues and limitations of energy storage approaches. Third, we shed light on the battery technologies, which are most frequently used in a wide range of applications for energy storage. The usage and types of batteries are described alongside their market shares and social and environmental aspects. Moreover, the recent advances in battery state estimation and cell-balancing mechanisms are reviewed.

State Estimation Models of Lithium-Ion Batteries for Battery Management System: Status, Challenges, and Future Trends

The state estimation technology of lithium-ion batteries is one of the core functions elements of the battery management system (BMS), and it is an academic hotspot related to the functionality and safety of the battery for electric vehicles. This paper comprehensively reviews the research status, technical challenges, and development trends of state estimation of lithium-ion batteries. First, the key issues and technical challenges of battery state estimation are summarized from three aspects of characteristics, models, and algorithms, and the technical challenges in state estimation are deeply analyzed. Second, four typical battery states (state of health, state of charge, state of energy, and state of power) and their joint estimation methods are reviewed, and feasible estimation frameworks are proposed, respectively. Finally, the development trends of state estimation are prospected. Advanced technologies such as artificial intelligence and cloud networking have further reshaped battery state estimation, bringing new methods to estimate the state of the battery under complex and extreme operating conditions. The research results provide a valuable reference for battery state estimation in the next-generation battery management system.

A three-time-scale dual extended Kalman filtering for parameter and state estimation of Li-ion battery

Estimation of battery state and parameters play an important role in electric vehicle battery management system (BMS). Second-order RC model is applied, the initial parameters of battery model are determined by experiments. Data points of open circuit voltage and state of charge (OCV-SOC) are determined by experiment. Different function forms are used to fit the OCV-SOC discrete points, and the function form with great fitting effect is selected as the OCV-SOC fitting form. Dual extended Kalman filter which is divided into Parameter filter and state filter is applied. Battery state in state filter is a fast-time-varying parameter, The battery model parameters in parameter filter are divided into two parts. the battery model parameters are classified according to the influence of each parameter on the terminal voltage. A longer sampling time is applied to the parameters that have a strong impact on the terminal voltage, and a longest sampling time is applied to the parameters that have a weak impact on the terminal voltage. The time-scale classification method is validated both quantitatively and qualitatively. Compared with the previous methods, the three-time-scale classification method can reduce the number of parameter updates.

A Novel Fusion Approach Consisting of GAN and State-of-Charge Estimator for Synthetic Battery Operation Data Generation

The recent success of machine learning has accelerated the development of data-driven lithium-ion battery state estimation and prediction. The lack of accessible battery operation data is one of the primary concerns with the data-driven approach. However, research on battery operation data augmentation is rare. When coping with data sparsity, one popular approach is to augment the dataset by producing synthetic data. In this paper, we propose a novel fusion method for synthetic battery operation data generation. It combines a generative, adversarial, network-based generation module and a state-of-charge estimator. The generation module generates battery operation features, namely the voltage, current, and temperature. The features are then fed into the state-of-charge estimator, which calculates the relevant state of charge. The results of the evaluation reveal that our method can produce synthetic data with distributions similar to the actual dataset and performs well in downstream tasks.