Showing papers on "State of charge published in 2020"
TL;DR: How machine learning methods and high-throughput experimentation provide a data-driven approach to this problem are discussed, and challenges in building models which provide fast and accurate battery state predictions are highlighted.
Abstract: Machine learning is a specific application of artificial intelligence that allows computers to learn and improve from data and experience via sets of algorithms, without the need for reprogramming. In the field of energy storage, machine learning has recently emerged as a promising modelling approach to determine the state of charge, state of health and remaining useful life of batteries. First, we review the two most studied types of battery models in the literature for battery state prediction: the equivalent circuit and physics-based models. Based on the current limitations of these models, we showcase the promise of various machine learning techniques for fast and accurate battery state prediction. Finally, we highlight the major challenges involved, especially in accurate modelling over length and time, performing in situ calculations and high-throughput data generation. Overall, this work provides insights into real-time, explainable machine learning for battery production, management and optimization in the future. Predicting the properties of batteries, such as their state of charge and remaining lifetime, is crucial for improving battery manufacturing, usage and optimisation for energy storage. The authors discuss how machine learning methods and high-throughput experimentation provide a data-driven approach to this problem, and highlight challenges in building models which provide fast and accurate battery state predictions.
285 citations
TL;DR: A cloud battery management system for battery systems to improve the computational power and data storage capability by cloud computing and a state-of-charge estimation algorithm with particle swarm optimization is innovatively exploited to monitor both capacity fade and power fade of the battery during aging.
Abstract: Battery management is critical to enhancing the safety, reliability, and performance of the battery systems This paper presents a cloud battery management system for battery systems to improve the computational power and data storage capability by cloud computing With the Internet of Things, all battery relevant data are measured and transmitted to the cloud seamlessly, building up the digital twin for the battery system, where battery diagnostic algorithms evaluate the data and open the window into battery’s charge and aging level The application of equivalent circuit models in the digital twin for battery systems is explored with the development of cloud-suited state-of-charge and state-of-health estimation approaches The proposed state-of-charge estimation with an adaptive extended H-infinity filter is robust and accurate for both lithium-ion and lead-acid batteries, even with a significant initialization error Furthermore, a state-of-health estimation algorithm with particle swarm optimization is innovatively exploited to monitor both capacity fade and power fade of the battery during aging The functionalities and stability of both hardware and software of the cloud battery management system are validated with prototypes under field operation and experimental validation for both stationary and mobile applications
260 citations
TL;DR: A data-driven method based on Gaussian process regression (GPR) is proposed to provide a feasible solution to SOC estimation of battery packs, and its superiority includes the ability to approximate nonlinearity accurately, nonparametric modeling, and probabilistic predictions.
204 citations
TL;DR: A survey of battery state estimation methods based on ML approaches such as feedforward neural networks, recurrent neural networks (RNNs), support vector machines (SVM), radial basis functions (RBF), and Hamming networks is provided.
Abstract: The growing interest and recent breakthroughs in artificial intelligence and machine learning (ML) have actively contributed to an increase in research and development of new methods to estimate the states of electrified vehicle batteries. Data-driven approaches, such as ML, are becoming more popular for estimating the state of charge (SOC) and state of health (SOH) due to greater availability of battery data and improved computing power capabilities. This paper provides a survey of battery state estimation methods based on ML approaches such as feedforward neural networks (FNNs), recurrent neural networks (RNNs), support vector machines (SVM), radial basis functions (RBF), and Hamming networks. Comparisons between methods are shown in terms of data quality, inputs and outputs, test conditions, battery types, and stated accuracy to give readers a bigger picture view of the ML landscape for SOC and SOH estimation. Additionally, to provide insight into how to best approach with the comparison of different neural network structures, an FNN and long short-term memory (LSTM) RNN are trained fifty times each for 3000 epochs. The error is somewhat different for each training repetition due to the random initial values of the trainable parameters, demonstrating that it is important to train networks multiple times to achieve the best result. Furthermore, it is recommended that when performing a comparison among estimation techniques such as those presented in this review paper, the compared networks should have a similar number of learnable parameters and be trained and tested with identical data. Otherwise, it is difficult to make a general conclusion regarding the quality of a given estimation technique.
191 citations
TL;DR: Experimental results illustrate that proposed ETNN-UKF can rapidly eliminate initial errors and provide satisfactory co-estimation performance, and a neural network is incorporated to enhance the performance of sub-models.
176 citations
TL;DR: Experimental results reveal that the proposed method can dramatically improve estimation accuracy compared with the solo LSTM method and the combined L STM-CKF method, and it exhibits excellent generalization ability for different datasets and convergence ability to address initial errors.
170 citations
TL;DR: A long short-term memory – recurrent neural network is proposed to model the sophisticated battery behaviors under varying temperatures and estimate battery SOC from voltage, current, and temperature variables and provides a satisfying SOC estimation under other temperatures which have no data trained before.
165 citations
TL;DR: This review critically investigates the various key implementation factors of the data-driven algorithms in terms of data preprocessing, hyperparameter adjustment, activation function, evaluation criteria, computational cost and robustness validation under uncertainties.
146 citations
TL;DR: A novel battery aging assessment method based on the incremental capacity analysis (ICA) and radial basis function neural network (RBFNN) model that provides insights for battery aging prediction based on massive real-time operation data.
Abstract: Accurate battery aging prediction is essential for ensuring efficient, reliable, and safe operation of battery systems in electric vehicle application. This article presents a novel battery aging assessment method based on the incremental capacity analysis (ICA) and radial basis function neural network (RBFNN) model. The RBFNN model is used to depict the relationship between battery aging level and its influencing factors based on real-world operation datasets of electric city transit buses. The ICA method together with the Gaussian window (GW) filter method is used to derive the peak values of IC curves which are utilized to represent battery aging levels, and the support vector regression (SVR) method is used in several scenarios for data preprocessing. The considered influencing factors include accumulated mileage of vehicles and initial charging state-of-charge (SOC), average charging temperature, average charging current, and average operating temperature of battery systems. The datasets collected from real-world electric city buses are used for RBFNN model training, validation, and test. The results show that an average prediction error of 4.00% is reached, and the derived model has a confidential interval of 92% with the prediction accuracy of 90%. This work provides insights for battery aging prediction based on massive real-time operation data.
144 citations
TL;DR: Simulation results verify that the momentum optimized GRU-RNN model can accurately and effectively estimate the SOC of the lithium battery.
130 citations
TL;DR: In this article, a hybrid pumped and battery storage (HPBS) system is proposed to make the off-grid renewable energy system more reliable and sustainable, considering the operating range of reversible pump-turbine machine, to extract maximum stored energy by operating HPBS at optimum efficiency.
TL;DR: The capacity of optimized machine learning techniques are presented toward enhanced SOC estimation in terms of learning capability, accuracy, generalization performance, and convergence speed and it is shown that the proposed method outperforms several state-of-the-art approaches in termsof accuracy, adaptability, and robustness under diverse operating conditions.
Abstract: State of charge (SOC) is a crucial index used in the assessment of electric vehicle (EV) battery storage systems. Thus, SOC estimation of lithium-ion batteries has been widely investigated because of their fast charging, long-life cycle, and high energy density characteristics. However, precise SOC assessment of lithium-ion batteries remains challenging because of their varying characteristics under different working environments. Machine learning techniques have been widely used to design an advanced SOC estimation method without the information of battery chemical reactions, battery models, internal properties, and additional filters. Here, the capacity of optimized machine learning techniques are presented toward enhanced SOC estimation in terms of learning capability, accuracy, generalization performance, and convergence speed. We validate the proposed method through lithium-ion battery experiments, EV drive cycles, temperature, noise, and aging effects. We show that the proposed method outperforms several state-of-the-art approaches in terms of accuracy, adaptability, and robustness under diverse operating conditions.
TL;DR: A dispatch interval coefficient is introduced to adjust the degree of conservatism, while improving the economy of microgrids system to solve the problem of over-conservatism of the robust optimization.
TL;DR: A framework for observation of the battery state-of-charge and remaining discharge time by using the unscented particle filter using the recursive method to predict the probable future current considering the historical data.
TL;DR: A novel co-estimation hierarchy for state of charge (SOC), state of health (SOH) and state of power (SOP) in lithium-ion batteries is devised and validated experimentally and presents remarkable benefits, compared to separate estimation solutions.
TL;DR: A stacked bidirectional long short-term memory (SBLSTM) neural network is proposed for SOC estimation and it is indicated that it can achieve good SOC estimation accuracy for different battery types at various ambient temperature conditions.
TL;DR: This paper proposes a joint lithium-ion battery state estimation approach that takes advantage of the data-driven least-square-support-vector-machine and model-based unscented-particle-filter and achieves the joint estimation with different time scales using the proposed hybrid joint state estimation method.
TL;DR: A novel BESS operational cost model considering degradation cost, based on depth of discharge and discharge rate is developed considering Lithium-ion batteries, and the approach can be applied to other conventional electrochemical batteries, but not flow batteries.
Abstract: Recent Federal Energy Regulatory Commission (FERC) Order 841 requires that Independent System Operators (ISOs) facilitate the participation of energy storage systems (ESSs) in energy, ancillary services, and capacity markets, by including ESS bidding parameters that represent the physical and operational characteristics. However, in the existing market frameworks that allow Battery Energy Storage Systems (BESSs) to participate, the bids and offers do not explicitly represent the physical and operational characteristics such as the state of charge (SOC), discharge rate, degradation, etc. This paper proposes a novel BESS operational cost model considering degradation cost, based on depth of discharge and discharge rate. The model is developed considering Lithium-ion batteries, and the approach can be applied to other conventional electrochemical batteries, but not flow batteries. A detailed bid/offer structure based on the proposed BESS operational cost functions is formulated. Thereafter, a new framework and mathematical model for BESS participation in an LMP based, co-optimized, energy and spinning reserve market, are developed. Three case studies are presented to investigate the impact of BESS participation on system operation and market settlement. The proposed model is validated on the IEEE Reliability Test System (RTS) to demonstrate its functionalities.
TL;DR: The results show that, over the full SOC range, the root-mean-square error of the battery pack SOC estimation is less than 0.6% and 1.5% using online and offline parameters, respectively.
TL;DR: A rapid acquisition method of the incremental capacity (IC) curve is established, then an adaptive capacity estimation framework based on ICA considering the charging condition is proposed, and the validation results show the effectiveness of the proposed correction method, which provides high accuracy and robustness.
TL;DR: A state observer for lithium-ion batteries based on an extended single-particle model is explored, which results in a trade-off between high accuracy and low computational burden, thus enables the real-time application.
TL;DR: In this paper, a model-free SOH calculation method by fusion of coulomb counting method and differential voltage analysis (DVA) is proposed to realize rapid online SOH estimation under constant current discharging stage.
TL;DR: A simplified model, in which the crucial parameters with high sensitivities are updated with SOC and SOH, while the other parameters retain their initial values, is proposed to ensure model accuracy while reducing computational complexity greatly.
TL;DR: In this article, the authors quantitatively analyzes the energy losses that take place during the charging of a battery electric vehicle (BEV), focusing especially in the previously unexplored 80% -100% State of Charge (SoC) area.
TL;DR: In this article, a real-time and approximately optimal energy management strategy based on Pontryagin's minimum principle (PMP) considering both fuel economy and power source durability is presented.
TL;DR: This article highlights the superiorities of the proposed leader–followers-based charging framework that combines the offline scheduling and online closed-loop regulation for battery pack charging, which brings benefits to significantly reduce the computational burden for the charger controller as well as improve the robustness to suppress the negative impact caused by the cell's model bias.
Abstract: Effective lithium-ion battery pack charging is of extreme importance for accelerating electric vehicle development. This article derives an optimal charging control strategy with a leader–followers framework for battery packs. Specifically, an optimal average state-of-charge (SOC) trajectory based on cells’ nominal model is first generated through a multiobjective optimization with consideration of both user demand and battery pack's energy loss. Then, a distributed charging strategy is proposed to make the cells’ SOCs follow the prescheduled trajectory, which can effectively suppress the violation of the safety-related charging constraints through online battery model bias compensation. This article highlights the superiorities of the proposed leader–followers-based charging framework that combines the offline scheduling and online closed-loop regulation for battery pack charging, which brings benefits to significantly reduce the computational burden for the charger controller as well as improve the robustness to suppress the negative impact caused by the cell's model bias. Extensive illustrative results demonstrate the effectiveness of the proposed optimal charging control strategy.
TL;DR: This study shows that a four-hidden-layer DNN trained on Dynamic Stress Test drive cycle is capable of predicting SOC values unexpectedly well of other unseen drive cycles such as Federal Urban Driving Schedule, Beijing Dynamic stress Test, and Supplemental Federal Test Procedure, respectively.
Abstract: The state of charge (SOC) is a crucial parameter of a battery management system for Li-ion batteries. The SOC indicates the amount of charge left in the battery of electric vehicles—akin to the fuel gauge in combustion vehicles. An accurate SOC knowledge contributes largely to the longevity, performance, and reliability of the battery. However, the SOC of Li-ion batteries cannot be easily measured by any apparatus. Furthermore, the SOC can also be influenced by numerous incalculable factors such as battery chemistry, ambient environment, aging factor, etc. In this article, we propose an SOC estimation model for a Li-ion battery using an improved deep neural network (DNN) approach for electric vehicle applications. We found that a DNN with a sufficient number of hidden layers is capable of predicting the SOC of the unseen drive cycles during training. We developed a series of DNN models with a varying number of hidden layers, and its training algorithm was to investigate their respective performance when evaluated on different drive cycles. We observe that the increasing number of hidden layers in the DNN (up to four hidden layers) decreases the error rate and improves SOC estimation. An additional increase in the number of hidden layers beyond that increases the error rate. In this study, we show that a four-hidden-layer DNN trained on Dynamic Stress Test drive cycle is capable of predicting SOC values unexpectedly well of other unseen drive cycles such as Federal Urban Driving Schedule, Beijing Dynamic Stress Test, and Supplemental Federal Test Procedure, respectively.
TL;DR: The main advantage of the proposed algorithms is eliminating the low-pass filter in the estimation algorithms, which can attenuate chattering that exists in the traditional sliding-mode observers and simplify the estimation algorithm.
Abstract: This paper investigates the real-time estimation on the state-of-charge (SoC) and state-of-health (SoH) of lithium-ion (Li-ion) batteries for the purpose of achieving reliable, safe, and efficient use of batteries. Three terminal sliding-mode observers (TSMOs) are designed; each observer is used to estimate one variable of a Li-ion cell for developing a real-time SoC estimation algorithm. To estimate the SoH, two additional TSMOs are subsequently presented. Finally, a set of complete estimation algorithms for SoC and SoH are formulated. The output injection signals of the proposed TSMOs are designed to be continuous. This can attenuate chattering that exists in the traditional sliding-mode observers and simplify the estimation algorithms. The main advantage of the proposed algorithms is eliminating the low-pass filter in the estimation algorithms. Therefore, higher estimation accuracy and faster response speed are obtained. The proposed methods are tested and evaluated using the acquired dynamic stress test and federal urban driving schedule test data, which demonstrate the effectiveness and feasibility.
TL;DR: Backward smoothing square root cubature Kalman filter (BS-SRCKF) is proposed to improve accuracy and convergence speed of SOC estimation and improved cuckoo search (ICS) algorithm is embedded in the standard particle filter (PF) to improve its performance.
TL;DR: The proposed EMS, called adaptive state machine strategy, employs the first two inputs to sort the FCSs out and the other inputs to do the power allocation, indicating promising improvement in the overall performance of the system.
Abstract: This paper aims at designing an online energy management strategy (EMS) for a multi-stack fuel cell hybrid electric vehicle (FCHEV) to enhance the fuel economy as well as the fuel cell stacks (FCSs) lifetime. In this respect, a two-layer strategy is proposed to share the power among four FCSs and a battery pack. The first layer (local to each FCS) is held solely responsible for constantly determining the real maximum power and efficiency of each stack since the operating conditions variation and ageing noticeably influence stacks’ performance. This layer is composed of a FCS semi-empirical model and a Kalman filter. The utilized filter updates the FCS model parameters to compensate for the FCSs’ performance drifts. The second layer (global management) is held accountable for splitting the power among components. This layer uses two inputs per each FCS, updated maximum power and efficiency, as well as the battery state of charge (SOC) and powertrain demanded power to perform the power sharing. The proposed EMS, called adaptive state machine strategy, employs the first two inputs to sort the FCSs out and the other inputs to do the power allocation. The ultimate results of the suggested strategy are compared with two commonly used power sharing methods, namely Daisy Chain and Equal Distribution. The results of the suggested EMS indicate promising improvement in the overall performance of the system. The performance validation is conducted on a developed test bench by means of hardware-in-the-loop (HIL) technique.