Kai Zhang

Differential comparison between faulty and normal cells is currently the main way to recognize internal short circuit (ISC) in the battery pack, which is prone to generate false alarms for inconsistent cells. In view of this, we innovatively propose a new method of ISC identification based on self-comparison of cell history data. The research data is the cell charging voltage with unified charging protocol after smoothing by wavelet transform method. The selection of research data is accomplished under the dynamic voltage window mechanism to ensure the robustness of the strategy to the different charging habits of the users. Based on the proposed voltage consistency assumption of adjacent charging cycles for one cell, the dynamic time warping algorithm is rationally chosen as the feature parameter extraction method to address the difficulty of similarity assessment for unequal length voltage series, which realizes the accumulation of voltage anomaly deviation. The improved density-based spatial clustering of applications with noise algorithm can realize the adaptive adjustment of hyperparameters according to the data discretization degree, which improves its adaptability to different datasets and avoids the misclassification of slight outlier samples. Based on experimental and real fault data, the final validation and comparison results indicate that the proposed strategy is robust to cell inconsistency and widely applicable to different types of batteries. An average accuracy of 99% and a one-time computation cost of 0.14 s are obtained, and potentially threatening hidden faults in real vehicles can be predicted up to 5 days in advance.

An accurate estimation of the state of health (SOH) of Li-ion batteries is critical for the efficient and safe operation of battery-powered systems. Traditional methods for SOH estimation, such as Coulomb counting, often struggle with sensitivity to measurement noise and time-consuming tests. This study addresses this issue by combining incremental capacity (IC) analysis and a novel neural network, Kolmogorov–Arnold Networks (KANs). Fifteen features were extracted from IC curves and a 2RC equivalent circuit model was used to identify the internal resistance of batteries. Recursive least squares were used to identify the parameters of the equivalent circuit model. IC features and internal resistance were considered as input variables to establish the SOH estimation model. Three commonly used machine learning methods (BP, LSTM, TCN) and two hybrid algorithms (LSTM-KAN and TCN-KAN) were used to establish the SOH estimation model. The performance of the five models was compared and analyzed. The results demonstrated that the hybrid models integrated with the KAN performed better than the conventional models, and the LSTM-KAN model had higher estimation accuracy than that of the other models. The model achieved a mean absolute error of less than 0.412% in SOH prediction in the test and validation dataset. The proposed model does not require complete charge and discharge data, which provides a promising tool for the accurate monitoring and fast detection of battery SOH.

Permanent magnet synchronous motors (PMSMs) are extensively utilized in production and manufacturing fields due to their wide speed range, high output torque, fast speed response, small size and light weight. PMSMs are susceptible to inter-turn short circuit faults, demagnetization faults, bearing faults, and other faults arising from irregular vibrations and frequent start–brake cycles. While fault diagnosis for PMSMs offers an effective means to enhance operational efficiency, the multi-sensor information fusion is often overlooked. In industrial production processes, the collected data inevitably suffers from noise contamination, which can adversely impact diagnostic outcomes. To enhance the robustness of diagnostic methods in noisy environments and mitigate the risk of overfitting, a PMSM fault diagnosis method based on image features of multi-sensor fusion is proposed. Firstly, the vibration acceleration signals of the PMSM at different positions were acquired. Then, the newly designed multi-signal Gramian Angular Difference Fields (MGADF) method combines sensor signals from three different installation locations into a single image. Next, the multi-texture features are fused to extract the features of the image. Various machine models are compared in the fault feature learning and classification, and the results show that the proposed diagnostic method has good diagnostic accuracy and robustness, with an average diagnostic accuracy of 99.54% and a standard deviation of accuracy of 0.19. It has excellent performance even in noisy environments. The method is non-invasive and can be extended and applied to the condition monitoring and diagnosis of industrial motors.

Energy management strategy (EMS) is a way to reduce the energy consumption of hybrid power systems. This article proposes a unique deep reinforcement learning- (DRL-) based EMS for plug-in hybrid electric heavy-duty trucks (PHETs), combining driving cycle pattern recognition (DPR) and deep transfer learning (DTL). The proposed EMS can cope well with the complex usage scenarios of PHETs and the difficulty of generating EMS. While ensuring the minimum overall driving cost, the strategy can improve the convergence speed of the DRL method and the generalizability under segmented usage scenarios. Firstly, representative driving cycles that reflect different usage scenarios are constructed based on a naturalistic data-driven method. Secondly, a plug-in hybrid electric heavy-duty truck (PHET) driving pattern recognizer based on a learning vector quantization neural network (LVQ) is built. Thirdly, the deep deterministic policy gradient (DDPG) algorithm is innovatively combined with the DTL algorithm. The pretrained neural network in the corresponding usage scenarios is transferred to the natural driving cycles based on DTL. Moreover, the proposed EMS gives an emphasized consciousness on the battery degradation cost. Finally, the strategy is tested under natural driving cycles in different usage scenarios and proven through comparison with the current state-of-the-art techniques, deep reinforcement learning-based strategy, and dynamic programming (DP). The results show that the proposed strategy outperforms existing cutting-edge deep reinforcement learning techniques in terms of convergence speed, battery life extension, fuel consumption, and overall driving cost reduction. The proposed control strategy can improve the convergence speed by nearly 50%, while effectively extending the battery life and reducing the overall driving cost compared to the existing state-of-the-art strategies. The battery degradation rate is reduced by 48.46%, 57.95%, and 36.99%, and the driving cost is reduced by 17.76%, 8.51%, and 7.12%, respectively, under each usage scenario.

The aim of this paper is to solve the problem for battery electric vehicles of low-precision and time-consuming inspection. A novel method of driving cycle development for battery electric vehicles’ operational safety is proposed in this paper. First, three inspection items are proposed based on relevant testing standards. The inspection calculation method of operational safety is developed based on the acceleration changing rate. Then the multi-cycle inspection method with the stable pedal mode is developed, and the Gauss filtering algorithm is applied for data preprocessing. A rapid inspection driving cycle construction method based on support vector machine is proposed, and a driving cycle is built with a total time of 204 s by fusing and splicing kinematic fragments. Finally, the proposed inspection calculation method is used to validate the operational safety inspection items by tracking the established rapid inspection driving cycle based on the test bench. The results shown are those that qualified the range of acceleration changing rate for driving stability [−0.35, −0.04]. The range for gliding smoothness is [0.05, 0.09]. The range for braking coordination is [−0.04, 0.095]. The maximum RMSE between the constructed rapid inspection segments is 9%, and the maximum RMSE between the tested driving segments is 6%. Test results meet design requirements. The thresholds for operational safety inspection items are evaluated based on the test results. We set less than 0.5 as the safety threshold for driving stability. During the experiment, gliding was less than 0.1 as the safety threshold for gliding comfort, and during braking it was less than 0.1 as the safety threshold for vehicle braking coordination.

Accurate estimation of vehicle states is extremely crucial for vehicle stability control. As a reliable estimation methodology, the unscented Kalman filter (UKF) has been widely utilized in vehicle control. However, the estimation accuracy still needs to be improved caused by the unpredictable measurement and process noise. In this paper, a novel modified UKF state estimation methodology combined with the ant lion optimization (ALO) is proposed for the stability control of a four in-wheel motor independent drive electric vehicle (4WIDEV). First, the optimal performance of the ALO algorithm is analyzed, where both unimodal and multimodal optimization test functions are selected and optimized by GA, PSO, and ALO, respectively. The results indicate that the ALO algorithm has good global optimization capability and applicability. Second, the ALO algorithm is merged into the UKF to adjust the statistical properties of noise information for the ALOUKF estimator design without extra sensor signals. At last, the simulations on the Matlab/Simulink-CarSim co-simulation platform and the road test based on an A&D 5435 rapid prototyping experiment platform (RPP) are carried out to verify the proposed method. The simulation and experiment results demonstrate that the ALOUKF estimator can improve state estimation accuracy and resist the vehicle nonlinearity even in the case of the complicated and emergency maneuvers.