To enhance the performance and reliability of power modules, the paper addresses inherent electro-thermal-mechanical multi-physics coupling characteristics. Utilizing Finite Element Analysis (FEA), comprehensive multi-physics simulations are conducted employing ANSYS software tools, including Q3D Extractor, Fluent, Maxwell, and Twin Builder. The simulation results demonstrate that parasitic inductance and thermal resistance significantly impact the switching characteristics and thermal management performance of the power modules. A thorough system-level evaluation is performed through thermal simulation, parasitic parameter extraction, and Double-Pulse Testing (DPT) simulations. Furthermore, the simulation accuracy is significantly improved by implementing an iterative verification process where experimental measurements are used to recalibrate the simulation models. This refined methodology provides a valuable reference for the subsequent optimization of power module design.

In order to address the issues of complex control strategy design and hardware circuit implementation of High Frequency AC (HFAC) resonant inverter power supply in Electric Vehicle (EV), this paper proposes a composite control strategy based on the combination of an integral controller and state feedback. Taking the typical LCLC DC-HFAC inverter as the research object, the Linear Quadratic Regulator (LQR) optimization control theory is used to realize the offline digital calculation of the feedback control parameters in the composite control strategy, which improves the dynamic performance of the DC/HFAC inverter and enhances the stability of the DC-HFAC inverter power supply. The control strategy and hardware circuit design are optimized by simplifying the parameter design process of the controller and the Phase-Shift Modulation (PSM) method. The experimental results show that the proposed LCLC DC-HFAC inverter power supply based on LQR optimized feedback composite control strategy not only has good steady-state performance, but also has high conversion efficiency and superior dynamic response speed.

In order To evaluate and predict the Electromagnetic Compatibility (EMC) performance of DC/DC converter in the early stage of design, the mainstream electromagnetic compatibility “three elements” method is first used to analyze the main interference source and propagation path of DC/DC converter. Secondly, based on the high-frequency parameter theory of transformer, the parasitic parameter theory of Printed-Circuit Board (PCB) and the parameter extraction method of common mode chokes, the common mode interference of transformer, PCB wiring and common mode chokes are analyzed separately. The high-frequency equivalent model of transformer and PCB, experimental environment test benches and high-voltage filtering modules are established using Maxwell, HFSS, SIwave, and Q3D software in the ANSYS simulation platform. Finally, the integration of each module of DC/DC converter and the simulation analysis of conduction and radiation emission are completed in Simplorer software. The results indicate that the conducted and radiated interference exceeds the standard more severely in the Metal-Oxide-Semiconductor Field Effect Transistor (MOSFET) switching frequency band and its harmonics of the main interference source. The model simulation results are basically consistent with theoretical analysis and actual experimental results, and the simulation model has high accuracy.

This paper investigates discrepancies between the conducted emission test results of the On-Board Charger (OBC) and those of the vehicle-level alternating current charging system. Starting from the testing mechanisms, the paper systematically analyzes the correlation between the OBC’s electromagnetic interference characteristics and the vehicle-level test conditions. Through combined simulation and experimental validation, the paper proposes a component-level conducted emission interference control scheme. By ensuring component-level electromagnetic compatibility performance, the scheme enables pre-validation of vehicle-level standard requirements, thereby provides support for the forward development of electromagnetic compatibility in new energy vehicles.

In order to meet the needs of charging safety, service experience of high-power DC charging piles and improve their power utilization, this paper proposes a flexible power allocation control strategy. Based on the topology of circular power allocation, the power allocation control timing and algorithm for charging start, charging in progress and release at the end are designed. The utilization rate of power nodes is improved by static and dynamic polling switching. To ensure stable operation of the system, the definition of minimum remaining required power is introduced, and the difference in remaining required power, the number of switching times in a single insertion gun, and the filtering time are comprehensively judged to avoid frequent switching. Verification result shows that this strategy can improve average power utilization rate from 1.76% to 2.24%, demonstrating significant optimization effect.

To predict the Remaining Useful Life (RUL) of Proton Exchange Membrane Fuel Cell (PEMFC) precisely, the paper proposes a method for predicting the RUL based on neural network optimized by Improved Snow Ablation Optimizer (ISAO). Firstly the original data are preprocessed by using Pauta criterion and wavelets, then the Pearson’s correlation coefficients are used to select parameters which have strong correlation with voltage as input variables. ISAO is used to optimize hyperparameters of Convolutional Neural Network-Gated Recurrent Unit (CNN-GRU) model. Then the CNN-GRU model is used to predict the output voltage of the PEMFC. Test results show that when the training set ratio is 30%, the mean absolute error is 1.6 mV, the root mean square error is 2.2 mV, the relative error is 0.41%, and the R-squared of the method is 99.20%, which are the best results the of six models. Compared with the Sparrow Search Algorithm (SSA), Snow Ablation Optimizer (SAO) and Whale Optimization Algorithm (WOA), the ISAO has faster optimization speed and better result, proving that the prediction model and the improved algorithm are effective.

The noise source and the noise transmission of the compressor NVH problem of the electric vehicle heat pump system are studied, and the improvement is made by optimizing the internal structure of the electric compressor, increasing the acoustic package and optimizing the air conditioning pipeline. The NVH test results show that the internal structure optimization can reduce the vibration excitation noise of the electric compressor while the addition of acoustic package and air conditioning pipeline optimization can reduce the compressor noise transmission. These enhancements contribute to a reduction in the noise level of the heat pump system during operation and the NVH performance of the vehicle.