查看更多>>摘要：The electric power supply on marine ships depends on the generator. Once the generator fails, it affects the operation of the marine ships. Stator interturn short circuit (SISC) is a common electrical fault in generators. In the case of a short circuit fault, the generator will be in the reinforced exciting current (REC) mode. At present, most research results mainly focus on fault detection methods, rarely concentrating on rotor temperature properties. In this article, the air gap magnetic flux density (MFD) model in the combination of different loads and SISC degrees is established, and the characteristics of the terminal voltage are presented. Moreover, the change rule of the stator current, MFD, and the loss of rotor are comprehensively investigated while considering REC mode. To reveal the law of the rotor thermal response and the biggest value of the deformation/stress/strain at the rotor core acted by temperature, the multiphysics field analysis model with electromagnetic-temperature-structure coupling is established. Experiments are carried out to verify the theoretical and finite element analysis (FEA) correctness by a 5 kW prototype generator.
查看更多>>摘要：Mechanical abuse is a cause of thermal runaway (TR) for lithium-ion batteries (LIBs). Developing an effective method to predict TR for LIBs under mechanical abuse is crucial for improving safety, which can enable drivers to get out of the car before the TR. However, TR under different mechanical conditions is complex and uncertain, and TR behavior prediction of LIBs coupled with mechanical–electrical–thermal is challenging. Therefore, an experimental platform was designed for conducting a series of mechanical abuse experiments for LIBs, and a mechanical–electrical–thermal coupling model for LIBs was established to supplement the experimental data for model training. Then, a data-driven model named TR temperature prediction neural network (TRTPNN) was constructed to decompose the battery prediction task into normal and failure states by setting a model-switching strategy, which can extract deeper information about the characteristic variables in different stages. The proposed model reduces an average mean absolute error (MAE) and mean absolute percentage error (MAPE) in different experimental conditions compared to the single neural network model by 17.6% and 41.7%, respectively. Finally, a multistage TR warning strategy based on the TRTPNN model is presented, triggering three alarm levels before the TR in all tested batteries.
查看更多>>摘要：Axial flux machines are increasingly attracting attention due to their power density and compact structure, making them particularly suitable for extended-range electric vehicles (EREVs). Among these, the coreless axial flux permanent magnet synchronous machine (CAFPMSM) with flat wire is renowned for its high copper filling, thermal conductivity, and efficiency. However, the flat wire in CAFPMSM is subject to substantial eddy current losses induced by the magnetic field. To mitigate eddy current loss, an unequal-thickness winding structure with multilayer flat wire is proposed in this article. A mathematical model is first developed to quantify the eddy current losses based on the air-gap flux density distribution. The relationship between the thickness of each layer of the winding and the resulting eddy current loss is analyzed. Second, the proposed winding configuration is further evaluated through 3-D finite element analysis (FEA) to assess its effectiveness in reducing losses. Finally, comprehensive experiments are carried out to demonstrate the effectiveness of the unequal-thickness windings in reducing eddy current loss and improving the efficiency of the machine.
查看更多>>摘要：In this article, a robust optimization design method is presented to improve the energy management effect of plug-in hybrid electric buses (PHEBs). Various uncertain factors are taken into account, including passenger load, resistance, and efficiency. First, the deterministic design of the energy management strategy is conducted under a city bus route, which is divided into 20 segments according to bus stations. The segmented equivalent consumption minimization strategy (ECMS) is established, wherein the equivalent factors (EFs) undergo optimization by the dynamic programming (DP) algorithm. Then, the sequential Taguchi method is utilized to optimize the EFs based on deterministic results. Uncertain factors are designated as noise factors, while EFs serve as control factors. The total fuel consumption is chosen as the optimization objective, with consideration given to the final state of charge (SOC) limit. The simulation results demonstrate that the energy management system obtained by robust optimization achieves a 1.9% reduction in fuel consumption expectation compared to the deterministic optimization. The result proves the validity of the proposed robust optimization method.
查看更多>>摘要：The stability of whirling modes of high-speed magnetically suspended rotors due to strong gyroscope effects varies with the speed, which is the main factor affecting the stability of the system. In this article, a speed-dependent linear parameter-varying (LPV) model of an active magnetic bearing (AMB) rotor system is derived. Then, an improved tensor product model transformation (TPMT) method is introduced to transform the dynamic model into a tensor product model form on a bounded parameter domain, in which the decomposed vertex systems are further remapped to analyze the stability easily. Moreover, a stability criterion based on dual-frequency Bode diagrams is developed. By analyzing the whirling stability characteristics of each vertex system, the stability judgment and stability margin calculation of the high-speed AMB rotor in the whole speed range can be realized, which effectively reduces the complexity of the radial rotation stability analysis of high-speed magnetically suspended rotors. The results show that this method is a feasible solution to analyze the stability of the parameter-dependent time-varying AMB model.
查看更多>>摘要：To achieve independent or refined control of modular wireless power transfer (MWPT) systems, precise mutual inductances are crucial. However, in MWPT systems, the number of mutual inductances grows factorially with the increase in the number of modules, making traditional single-channel mutual inductance calculation methods computationally complex and inefficient in practical applications. To address these challenges, this article introduces an offline mutual inductance estimation methodology for MWPT systems with a series-series (SS) compensation topology. By utilizing blocking or bypass switches in each module, measurement topologies capable of calculating multiple mutual inductances efficiently are constructed. To obtain sufficient equations, multiple measurements at different operating frequencies are performed. This approach effectively mitigates the estimation inaccuracy stemming from the tolerances of resonant components by considering the resonant frequency as a variable to be solved. Furthermore, due to the challenges and potential significant errors associated with phase angle measurement under high-frequency operating conditions, a phase identification method is introduced so that the proposed methodology only relies on measuring amplitudes of winding currents and voltages. Moreover, a quantitative analysis of the factors that may cause errors in the estimation is provided. Finally, two-module and four-module WPT prototypes are constructed to validate the proposed method.
查看更多>>摘要：Accurate and reliable health status prognostics are critical for ensuring battery safety and enabling smart management. Most existing studies assume stable operating conditions, which contrast with the dynamic and uncertain characteristics of in-service electric vehicles (EVs), thus challenging the practical application of developed approaches. To address this challenge, we propose an adaptable battery degradation prediction framework for EVs with different operating characteristics. Initially, we analyze the operational characteristics of EVs across different application scenarios and introduce a cluster-based charging pattern identification approach. Subsequently, we perform targeted feature extraction based on the identified charging patterns and propose a multilevel feature selection strategy to construct a comprehensive and effective feature pool. Furthermore, we develop two neural network (NN)-based models for reconstructing historical capacity trajectories and predicting battery degradation, further integrating transfer learning to enhance model efficiency and accuracy in unknown scenarios. Finally, we validate the proposed battery health prognostic framework across various training and prediction scenarios, demonstrating its high accuracy and reliability. Specifically, the mean absolute percentage error (MAPE) and root mean square error (RMSE) of degradation prediction are found to be within 1.30% and 2.05% for EVs in different operating scenarios, representing a notable improvement to existing methods.
NETL
NSTL
IEEE