Remaining useful life (RUL) prediction of lithium-ion batteries can assess the battery reliability to determine the advent of failure and mitigate battery risk. The existing RUL prediction techniques for lithium-ion batteries are inefficient for learning the long-term dependencies among the capacity degradations. This paper investigates deep-learning-enabled battery RUL prediction. The long short-term memory (LSTM) recurrent neural network (RNN) is employed to learn the long-term dependencies among the degraded capacities of lithium-ion batteries. The LSTM RNN is adaptively optimized using the resilient mean square back-propagation method, and a dropout technique is used to address the overfitting problem. The developed LSTM RNN is able to capture the underlying long-term dependencies among the degraded capacities and construct an explicitly capacity-oriented RUL predictor, whose long-term learning performance is contrasted to the support vector machine model, the particle filter model, and the simple RNN model. Monte Carlo simulation is combined to generate a probabilistic RUL prediction. Experimental data from multiple lithium-ion cells at two different temperatures are deployed for model construction, verification, and comparison. The developed method is able to predict the battery's RUL independent of offline training data, and when some offline data is available, the RUL can be predicted earlier than in the traditional methods.

To improve the use of lithium-ion batteries in electric vehicle (EV) applications, evaluations and comparisons of different equivalent circuit models are presented in this paper. Based on an analysis of the traditional lithium-ion battery equivalent circuit models such as the Rint, RC, Thevenin and PNGV models, an improved Thevenin model, named dual polarization (DP) model, is put forward by adding an extra RC to simulate the electrochemical polarization and concentration polarization separately. The model parameters are identified with a genetic algorithm, which is used to find the optimal time constant of the model, and the experimental data from a Hybrid Pulse Power Characterization (HPPC) test on a LiMn₂O₄ battery module. Evaluations on the five models are carried out from the point of view of the dynamic performance and the state of charge (SoC) estimation. The dynamic performances of the five models are obtained by conducting the Dynamic Stress Test (DST) and the accuracy of SoC estimation with the Robust Extended Kalman Filter (REKF) approach is determined by performing a Federal Urban Driving Schedules (FUDS) experiment. By comparison, the DP model has the best dynamic performance and provides the most accurate SoC estimation. Finally, sensitivity of the different SoC initial values is investigated based on the accuracy of SoC estimation with the REKF approach based on the DP model. It is clear that the errors resulting from the SoC initial value are significantly reduced and the true SoC is convergent within an acceptable error.

Battery technology is the bottleneck of the electric vehicles (EVs). It is important, both in theory and practical application, to do research on the modeling and state estimation of batteries, which is essential to optimizing energy management, extending the life cycle, reducing cost, and safeguarding the safe application of batteries in EVs. However, the batteries, with strong time-variables and nonlinear characteristics, are further influenced by such random factors such as driving loads, operational conditions, in the application of EVs. The real-time, accurate estimation of their state is challenging. The classification of the estimation methodologies for estimating state-of-charge (SoC) of battery focusing with the estimation method/algorithm, advantages, drawbacks, and estimation error are systematically and separately discussed. Especially for the battery packs existing of the inevitable inconsistency in cell capacity, resistance and voltage, the advanced characterizing monomer selection, and bias correction-based method has been described and discussed. The review also presents the key feedback factors that are indispensable for accurate estimation of battery SoC, it will be helpful for ensuring the SoC estimation accuracy. It will be very helpful for choosing an appropriate method to develop a reliable and safe battery management system and energy management strategy of the EVs. Finally, the paper also highlights a number of key factors and challenges, and presents the possible recommendations for the development of next generation of smart SoC estimation and battery management systems for electric vehicles and battery energy storage system.

An adaptive Kalman filter algorithm is adopted to estimate the state of charge (SOC) of a lithium-ion battery for application in electric vehicles (EVs). Generally, the Kalman filter algorithm is selected to dynamically estimate the SOC. However, it easily causes divergence due to the uncertainty of the battery model and system noise. To obtain a better convergent and robust result, an adaptive Kalman filter algorithm that can greatly improve the dependence of the traditional filter algorithm on the battery model is employed. In this paper, the typical characteristics of the lithium-ion battery are analyzed by experiment, such as hysteresis, polarization, Coulomb efficiency, etc. In addition, an improved Thevenin battery model is achieved by adding an extra RC branch to the Thevenin model, and model parameters are identified by using the extended Kalman filter (EKF) algorithm. Further, an adaptive EKF (AEKF) algorithm is adopted to the SOC estimation of the lithium-ion battery. Finally, the proposed method is evaluated by experiments with federal urban driving schedules. The proposed SOC estimation using AEKF is more accurate and reliable than that using EKF. The comparison shows that the maximum SOC estimation error decreases from 14.96% to 2.54% and that the mean SOC estimation error reduces from 3.19% to 1.06%.