The common-mode(CM) inductor is composed of windings and a magnetic core. Mn-Zn ferrite can be used as the magnetic core of CM inductor, and the frequency-dependent characteristics of its high-frequency parameters make the CM inductor exhibit non-linearity within a range of 150 kHz-30 MHz. It is difficult to accurately predict the impedance characteristics of the CM inductor in a high-frequency band due to the large deviation between the traditional lumped circuit model of CM inductor and the measured impedance characteristics. The effects of material parameters of the magnetic core and the winding scheme on the high-frequency impedance characteristics of CM inductor are analyzed, and the mechanism of magnetic material parameters and the winding structure affecting the distributed capacitance is described from the perspective of the distributed capacitance characteristics of CM inductor. A wide-band impedance simulation modeling method for CM inductor with the consideration of related frequency-dependent parameters of magnetic core was proposed, and its validity was verified by combining with the impedance test result.

With the increasing attention to environmental issues, more and more distributed energy systems represented by microgrids are appearing in the power system, which also poses some challenges to the traditional power systems. For example, the delay in digital control system, variations in grid impedance of weak grid and the interaction between parallel converters in microgrids will cause adverse effects on the stable operation of microgrids. On this basis, a novel type of grid-forming control method for microgrid considering control delay and variation in grid impedance is put forward to enhance the stability of microgrids under uncertainties. First, the above problems are modeled, and a delay compensation method is proposed to improve the robustness of the control system with respect to the variation in grid impedance. Then, a feedforward loop is introduced into the control system to protect it from the interference of parallel inverters in microgrids. Finally, experimental results demonstrate the effectiveness and superiority of the proposed control method.

Aimed at the problem of wide frequency range and large circulating current with the traditional frequency-controlled LLC resonant converter in wide output voltage applications, a fixed-frequency PWM controlled hybrid bridge dual-LLC resonant converter is studied. According to the difference in the primary-side structure, the converter has three forms of topology, i.e., half-bridge-half-bridge, half-bridge-full-bridge and full-bridge-full-bridge, in which the primary-side structure is in parallel and the two transformers on the secondary-side are in series. Compared with the traditional frequency-controlled LLC converter, the three topologies always work at the resonant frequency, which reduces the switching frequency range. In addition, under the PWM control strategy, the three topologies can achieve 2, 3 and 4 times voltage gain, respectively, thereby adapting to wide voltage scenarios. At the same time, the circuit has a low circulating current loss and a good soft switching performance. Simulink simulation and experimental results verified the feasibility of the proposed scheme.

For evaluating the capacity of wind powers, photovoltaics and other power electronic grid-connected units supporting power systems, the core foundation is to correctly understand the unit's functional role (i.e., the unit characteristics) that unit adjusts its own internal voltage amplitude/frequency according to the active/reactive power imbalance. However, the mainstream PLL-based grid-connection structure in power electronic units seriously hinders the understanding of the unit's functional role. In particular, based on a specific PLL-based grid-connection structure, the industry and academia at present form a "grid following" role perception that the internal voltage of unit follows the grid voltage or terminal voltage, and have not recognized the functional role that the unit should take during the system operation. Therefore, through an in-depth understanding of the independent excitation-response mechanism of current control which is hidden under the PLL-based grid-connection structure, i.e., the internal voltage response depends on current excitation alone, the functional role of PLL-based grid-connected units in which the active/reactive power imbalance independently adjusts the internal voltage amplitude/frequency is clarified. Afterwards, a role characterization method for unit is proposed based on the relationship between active/reactive power imbalance excitation and internal voltage amplitude/frequency response, i.e., the amplitude-frequency motion equation. Finally, the inevitability of characterizing the role of PLL-based grid-connected units through the relationship between power excitation and internal voltage response is elaborated on, and the existing limitations in the understanding of the role of PLL-based grid-connected units in industry and academia are pointed out.

To improve the state-of-charge(SOC) prediction accuracy of lithium battery, a prediction method based on the fusion model of Attention mechanism and convolution neural network-long short-term memory(CNN-LSTM) is proposed. This model uses one-dimensional CNN and LSTM neural network to learn the nonlinear relationship between SOC and lithium battery discharge data, as well as the long-term dependence existing in SOC sequences. At the same time, it adopts a "many-to-one" structure and establishes a mapping relationship between the SOC at the present moment and the discharge data at multiple historical moments, and pays attention to the historical discharge data which has a greater influence on the SOC at the present moment through the Attention mechanism, thus further improving the SOC prediction accuracy. The SOC prediction experiments under dynamic conditions show that the average prediction error of the proposed method is 0.89% under different temperature conditions, which is 81.2%, 66.7% and 56.5% lower than those of SVM, GRU and XGBoost algorithms, respectively. In addition, this method is also superior to LSTM and CNN-LSTM models that do not combine the Attention mechanism, showing a higher prediction accuracy and higher application values.

To solve the problems of large current stress, difficult soft switching of all switches and slow dynamic response of dual active bridge (DAB) converters, a multi-objective unified optimal control strategy based on triple-phase-shift(TPS) control is proposed. The forward power flow global mode of TPS control is analyzed, and three high-efficiency modes are selected to establish the analytical models of current stress and soft switching. Combined with these models, the optimal phase-shift ratio combination and minimum current stress in different modes are derived using the cost function optimization equation, which makes the switches operate within the zero-voltage-switching power constraint range. At the same time, the virtual power component is introduced in the process of efficiency optimization. A small-signal model is constructed, and the influence of small disturbance of different state variables on output voltage is clarified. Experimental results show that the proposed control strategy can not only reduce the current stress of the DAB converter and make all switches realize zero-voltage-switching, but also improve the dynamic performance of output voltage in the full power range.

To ensure the safe, reliable and economic operation of high-speed railway, a multi-dimensional control method for the power supply operation energy of high-speed railway cophase power supply system is proposed. The composition structure and power supply process of the power supply system are analyzed, and a power flow controller is used to compensate the power of the system, thus ensuring its stable operation. Combined with indicators such as three-phase imbalance degree and distortion of voltage and current waveforms, the optimal load model of operation energy is constructed to minimize the loss of transmission energy. Through the active and reactive power compensation for current in two power supply arms, the negative-sequence current is eliminated. A control strategy for the railway power conditioner is formulated, and the changes in the step-down transformer are obtained to ensure a stable DC-side voltage. Afterwards, a proportional integral controller is used to obtain the ideal value of current and get the control signal, thus realizing the multi-dimensional control of power supply operation energy. Simulation results show that the proposed multi-dimensional control method can improve the three-phase current balance degree and achieve an effect of voltage sharing stability.

To improve the efficiency of a dual-active-bridge (DAB) converter, a control strategy of minimum current stress with varying switching frequency control is proposed. First, the minimum-current-stress-optimized control method when the voltage changes in a wide range is analyzed, the expressions for the conduction loss and switching loss of the DAB converter under light load conditions are established, and the conduction loss and switching loss at different switching frequencies are further compared. Then, the implementation method for closed-loop control and the power transmission range in each mode are introduced in detail. Under light load conditions, the proposed method can significantly reduce the current stress while improving the efficiency of the DAB converter. Finally, simulation and experimental results verified the advantage of the proposed control method.

With the development of industrial demand, higher requirements are put forward for the reliability of power transistors. The real-time measurement of device junction temperature can ensure the normal operation of the device, so it is very important. Through the research on the on-line measurement method for turn-on delay time, a measurement circuit is designed. By measuring the turn-on delay time at different junction temperatures, the relationship between them is established, the on-line measurement of device junction temperature is realized by measuring the turn-on delay time under actual working conditions, and the device junction temperature is deduced. The results measured using the proposed method are similar to those obtained by an infrared temperature gun. In this way, an on-line measurement method for MOSFET junction temperature based on turn-on delay time is proposed, providing a new idea for the on-line measurement of junction temperature in the future.

Along with the widespread applications of lithium batteries in industry and daily life, the efficiency and speed under balanced charging strategies for lithium battery packs have become increasingly important. A modular cell-to-pack-to-cell(CPC) balanced charging system is constructed to solve the problem of fast equalization charging for lithium battery packs. First, the equalization system is modularized, and the equalization circuits within and between modules are established. Then, an optimization strategy for balanced charging is proposed, under which the proposed model is solved hierarchically, i.e., the charging time is calculated using the binary method in the top layer, and the charging current is optimized using the gradient descent method in the bottom layer. Finally, through a comparison with the charging time, equalization time, cell terminal voltage and equalizer voltage under the non-modular balanced charging strategy, the feasibility and effectiveness of the proposed strategy are verified.