Figure 18 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
The state-of-energy (SOE) and state-of-health (SOH) are two crucial quotas in the battery management systems, whose accurate estimation is facing challenges by electric vehicles’ (EVs) complexity and changeable external environment. Although the machine learning algorithm can significantly improve the accuracy of battery estimation, it cannot be pe...
Citations
... In this method, a learning-based capacity estimation model is first created using a gated convolutional and bi-directional long short-term memory hybrid neural network (GateCNN-BiLSTM). The CNN is used to capture the advanced spatial features of the battery sequence [29], and the bi-directional long short-term memory (BiLSTM) neural network is used to extract the positive and negative context information of the battery sequence [30]. The gating mechanism can suppress noise information and enhance the sensitivity of the network to important features, thereby improving the generalization ability of the network. ...
Accurately estimating the capacity degradation of lithium-ion batteries (LIBs) is crucial for evaluating the status of battery health. However, existing data-driven battery state estimation methods suffer from fixed input structures, high dependence on data quality, and limitations in scenarios where only early charge–discharge cycle data are available. To address these challenges, we propose a capacity degradation estimation method that utilizes shorter charging segments for multiple battery types. A learning-based model called GateCNN-BiLSTM is developed. To improve the accuracy of the basic model in small-sample scenarios, we integrate a single-source domain feature transfer learning framework based on maximum mean difference (MMD) and a multi-source domain framework using the meta-learning MAML algorithm. We validate the proposed algorithm using various LIB cell and battery pack datasets. Comparing the results with other models, we find that the GateCNN-BiLSTM algorithm achieves the lowest root mean square error (RMSE) and mean absolute error (MAE) for cell charging capacity estimation, and can accurately estimate battery capacity degradation based on actual charging data from electric vehicles. Moreover, the proposed method exhibits low dependence on the size of the dataset, improving the accuracy of capacity degradation estimation for multi-type batteries with limited data.
... SOE is one of the basic parameters of the battery safety protection module in BMS, which directly reflects the internal energy change of the battery [33]. Under the current situation where the range estimation accuracy is not high, SOE is used to achieve the range estimation of new energy vehicles and provide the driver with accurate range information [34][35][36]. Accurate SOE estimation not only helps the battery management system to develop a reasonable energy control strategy and optimize the energy control performance of new energy electric vehicles but also has practical significance to promote the development and promotion of electric vehicles [37][38][39][40][41]. The classification of the SOE estimation method is roughly the same as that of the SOC estimation method, which can be divided into the direct method, model-based method, and data-driven method [42][43][44]. ...
Accurate online estimation of the state of charge (SOC) and state of energy (SOE) of lithium-ion batteries are essential for efficient and reliable energy management of new energy electric vehicles (EVs). To improve the accuracy and stability of the joint estimation of SOC and SOE of lithium-ion batteries for EVs, based on a dual-polarization (DP) equivalent circuit model and time-varying forgetting factor recursive least squares (TVFFRLS) algorithm for online parameter identification, a joint estimation method based on singular value decomposition with adaptive embedded cubature Kalman filtering (SVD-AECKF) algorithm is proposed. The algorithm adopts the embedded cubature criterion and singular value decomposition method to improve filtering efficiency, accuracy, and numerical stability. Meanwhile, combining the idea of adaptive covariance matching for real-time adaptive updating of system noise to improve joint estimation accuracy. Finally, the results under different initial errors and complex operating conditions show that the SVD-AECKF algorithm improves the convergence time of SOC estimation by at least 26.3% compared to that before optimization. The SOE estimation error is reduced by at least 12.0% compared to that before optimization. This indicates that the SVD-AECKF algorithm has good joint SOC and SOE estimation accuracy, convergence, and stability.
... For non-real-time estimated states, such as SOE and SOH, the method of vehicle-cloud collaboration approach is still promising. Mei et al. [131] proposed a joint SOE and SOH prediction algorithm, which combines long short-term memory (LSTM), Bi-directional LSTM (Bi-LSTM), and convolutional neural networks (CNNs) for EVs based on vehicle-cloud collaboration, given in Figure 10. The SOH correction in SOE estimation achieves the joint estimation with different time scales. ...
... schematic diagram of vehicle-cloud collaboration for joint estimation[131]. ...
Remaining 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.
Accurate estimation of lithium‐ion battery state of energy (SOE) is an important prerequisite for prolonging battery life and ensuring battery safety. To achieve a high‐precision estimation of the SOE, while a ternary lithium‐ion battery being the specifically targeted in this study, a novel method for SOE estimation is proposed, which combines limited‐memory recursive least squares (LM‐RLS) with strong tracking adaptive window Multi‐innovation cubature Kalman filtering (STW‐MCKF). In the LM‐RLS algorithm, the model parameters at the current time are updated with a limited dataset to solve the data saturation problem and improve the recognition accuracy of the RLS algorithm. The CKF algorithm is optimized by the STW algorithm in the STW‐CKF algorithm to enhance its robustness under strong disturbances. Additionally, a self‐adaptive window multiple innovation strategy is proposed to improve the accuracy of SOE estimation and the stability of the CKF algorithm, while maintaining a balance between computational complexity and SOE estimation accuracy. To validate the effectiveness of the algorithm, experiments are conducted under DST and BBDST conditions. The results show that the STW‐MCKF algorithm has a maximum convergence time of 4 s and an SOE estimation error within 1.04% under DST conditions. Under BBDST conditions, the maximum convergence time is 3 s, and the SOE estimation error is within 2.34%. Furthermore, the STW‐MCKF algorithm demonstrates good stability under the two conditions, indicating the effectiveness of the proposed method for lithium‐ion battery SOE estimation.
State of charge and state of energy are essential performance indicators of the battery management system and the key to reflecting the remaining capacity of batteries. Aiming at the problems of low precision, long time, and strongly nonlinear system estimation of state of charge and state of energy of lithium‐ion batteries based on traditional algorithm under complex working conditions, this paper proposes a hybrid method consisting of the long short‐term memory neural network and square root extended Kalman smoothing. The long short‐term memory neural network can enhance the memory ability of the previous time data. The sliding window technology is introduced into the network to improve the correlation between the last time and the subsequent time estimation. Based on the traditional Kalman filtering algorithm, the square root and reverse smoothing algorithms are introduced to solve the risk of the negative covariance matrix and the problems of slow convergence and significant estimation deviation caused by a strongly nonlinear system. According to experiments, under the hybrid pulse power characterization working condition at 25°C, the maximum absolute errors of state of charge and state of energy are 1.779% and 1.487%, and the mean absolute errors are 0.352% and 0.894%, respectively. Under the Beijing bus dynamic stress test working condition at 25°C, the maximum absolute errors of state of charge and state of energy are 2.703% and 2.369%, and the mean absolute errors are 0.462% and 0.621%, respectively. The experimental results show that this algorithm can obtain reliable state of charge and state of energy under different complex working conditions with high accuracy, convergence, and robustness.
