Different winding schemes for a toroidal inductor can affect its magnetic field leakage, which will intensify the near-filed coupling of an electromagnetic interference (EMI) filter. First, the effect of equivalent radial current and tangential current on magnetic field leakage was studied by using the partial element equivalent circuit method, and it was found that the equivalent current loop composed of tangential current was the main source of axial magnetic field leakage in the toroidal inductor. Then, a winding scheme was proposed, in which a PCB was sandwiched by two magnetic cores to generate reverse current loops above and below the PCB for magnetic field cancellation. In the experimental link, three-dimensional finite element simulation was conducted to demonstrate that the proposed scheme can further reduce the axial magnetic field leakage while maintaining a similar inductance as the traditional scheme. The impedance and insertion loss tests of the EMI filter were conducted, and it was verified that compared with the traditional scheme, the novel scheme can weaken the influence of high-frequency parasitic parameters on the inductor’s impedance characteristic, thereby further optimizing the high-frequency filtering performance of the EMI filter.

The electromagnetic interference (EMI) filter is an effective approach to suppressing EMI. However, due to the influence of high-frequency parasitic parameters, EMI filters are difficult to effectively suppress high-frequency EMI noises. In this paper, the characteristics of high-frequency parasitic parameters of EMI filter components including common-mode (CM) inductor and X capacitor are analyzed, and a high-frequency model exhibiting the wide-band impedance characteristics of filter components is established. Moreover, the influence of frequency-varying permeability of ferrite cores on the parasitic parameter cancellation effect is analyzed. The extraction method for the equivalent parallel capacitance of the CM inductor is investigated, the selection of CM inductor cancellation capacitance is optimized, and the wide-band noise filtering of the CM inductor is realized. The wide-band noise filtering characteristics of the whole EMI filter are predicted by finite element simulation. Combined with the simulation and experimental test of insertion loss, the effectiveness of the proposed EMI filter frequency-domain modeling method and the selection basis of CM inductor cancellation capacitance considering the frequency variation of permeability was verified.

HIRFL-CSRe is an experimental ring part of the CSR system in the post-accelerated cooling storage ring of HIRFL in Lanzhou heavy-ion accelerator facility, in which the bipolar ferromagnet has higher requirements for the performance of the power supply. To meet the requirements of high-precision indicators for this kind of ferromagnet power supply and solve its technical difficulties, a high-precision DC power supply device is developed. The main circuit of the power supply adopts a multiplex circuit topology, the control part uses a multi-variable coupling digital-analog hybrid control strategy, and the process structure adopts standardized design for the power converter and other key components. Experimental results show that the output current stability of the power supply was less than or equal to 5×10^-6/8 h, and the other performance indicators of the power supply also met the physical design requirements for the accelerator. Finally, the power supply passed the acceptance test.

The collection of data from high-power power supply output signals (voltage/current) will be unavoidably interfered by electromagnetic noise, which affects the analysis of voltage/current signals. Therefore, the denoising of high-power power supply signals is an essential process. For the analysis of output signal characteristics of high-power power supply, a denoising method is proposed, in which a morphological filter is superimposed on a modified wavelet threshold function. The modified wavelet threshold denoising method can filter random noises by adjusting parameters, which can recover useful signals from noisy signals more effectively and reduce unnecessary distortion while taking into account the advantages of soft/ hard threshold functions. In comparison, the morphological filter is mainly used to filter impulse interference. Simulation results show that the proposed algorithm can effectively reduce noise in the high-power power supply signal, which largely preserves the characteristics of the actual signal. From the comparison between the proposed method and other modified wavelet algorithms on which the morphological filter is superimposed, it is found that the signal-to-noise ratio is improved by 17.5%. In addition, the superiority of the pro-posed method was verified by experimental results.

The traditional on-load voltage regulation technology has deficiencies such arc, a slow response speed and a low regulation accuracy, which affect the operation reliability of a transformer. Although the novel power electronic transformers have advantages of no arc and fast response, they have a large loss and a high cost. On the basis, a novel single-phase four-column magnetron transformer structure is proposed, which consists of a magnetron reactor and a transformer connected in series. According to the topological structure of the magnetron reactor, the working principle for the magnetron transformer is analyzed, and the circuit equation and electromagnetic equation are deduced. The right-side yoke in this structure can provide an AC flux loop, which is not interfered by the DC flux. The Maxwell and Simplorer in software Ansys are used to conduct field-circuit coupled multi-physics domain co-simulation to analyze the voltage regulation capability of the model, and the harmonic analysis of output voltage is carried out in Simulink. It is verified that the proposed novel magnetron transformer can adjust the output voltage smoothly without arc and without step, and the harmonic content is low.

DC solid-state circuit breakers (SSCBs) based on SiC MOSFET have attracted much attention owing to their capability of quickly breaking the fault currents. However, the failure mechanism of SiC MOSFET is more complex than those of traditional power semiconductor devices. SiC MOSFET has special gate failure and aging characteristics, but the relevant characteristics and mechanism are not clear, which brings great challenges to the failure and aging discrimination of SSCBs. A 400 V DC SSCB based on SiC MOSFET was taken as the research object, and the single and repetitive hard switching characteristics of the device under this working condition were investigated. It was verified that gate failure may occur in SiC MOSFET during the process of single or repetitive actions, and a new aging characteristic of gate voltage reduction appeared. Based on simulation tools, the temperature and thermal stress distribution inside the device during the process were obtained. It was found that the temperature rise near the gate structure of the device was close to 1 000 K, and a thermal stress of 1.4 GPa would be generated. Through the opening, slicing and SEM scanning of failed devices, the cracks in the gate oxide layer were confirmed, and the mechanism of gate failure and aging caused by gate structure cracking due to thermal stress was verified.

Abstract: The IGBT module of a motor controller is one of the important components for realizing electric energy conversion and motor control. This module usually works under complex conditions, and its reliability is crucial for the stability of the whole electric drive system. In general, the IGBT loss is directly calculated using the method of manual formulas, which ignores the influence of temperature rise on the IGBT module. Aimed at this problem, a method of thermal coupling analysis is proposed. First, a loss calculation model of IGBT module is established. On the basis, a thermal network model is built for the estimation of junction temperature, and the temperature rise estimation result is further used to correct the loss calculation model. A simulation model of the motor controller is constructed, and the junction temperature and loss of IGBT are simulated. In addition, the effectiveness of the proposed thermal coupling analysis method was verified by the results of a temperature rise experiment conducted on the motor controller.

The loss calculation accuracy of a three-phase full-bridge inverter will be reduced if the parasitic capacitance and parasitic inductance loss of SiC MOSFET in the switching process and the conduction loss caused by changes in the duty cycle in the conduction process are ignored. To solve the problems, a physical model of SiC MOSFET considering parasitic capacitance and parasitic inductance was proposed. The parasitic capacitance and parasitic inductance were obtained by querying a data sheet and using an oscillation method. The losses of parasitic inductance and parasitic capacitance during the switching process were calculated, the average conduction loss equation under space vector pulse width modulation (SVPWM) was deduced, and the three-phase full-bridge loss was obtained. Finally, experiments were carried out on the control system of a permanent magnet synchronous motor based on SiC MOSFET. The influence of PWM frequency on the three-phase full-bridge loss was analyzed, and the accuracy of parameters for the physical model of SiC MOSFET was verified. Moreover, the loss calculation method based on the physical model of SiC MOSFET can improve the loss calculation accuracy of the inverter.

The lithium-ion battery equalizer based on a Buck-Boost converter can only transfer the energy in a step-by-step way, resulting in a long energy transmission path and a low equalization efficiency. On this basis, a lithium-ion battery equalizer based on dual active half-bridge is studied. The key parameters affecting the energy flow between cells within a group and between those in different groups in the half-bridge equalizer are analyzed, and a phase-shift control strategy is put forward accordingly. The dual active half-bridge equalizer under phase-shift control can realize a fast energy transfer between cells in the half-bridge group at a high frequency. Since the equalizer operates in a high-frequency state, its volume and weight are greatly reduced. The expressions for the equivalent equalization current between cells within a group and that between cells in different groups were derived in detail, and a simulation model and an experimental prototype were built to perform a verification. Simulation and experimental results show that the proposed dual active half-bridge lithium-ion battery equalizer based on phase-shift control can quickly realize voltage equalization between battery cells.

The massive grid connection of renewable energy brings a series of problems to the power grid, such as an insufficient frequency modulation capacity of thermal power units and a poor frequency modulation effect. Aimed at these problems, a combined frequency modulation strategy for thermal power and energy storage based on a battery energy storage system is proposed. First, the frequency modulation signal of area control error (ACE) is decomposed using variational mode decomposition (VMD). Combined with the load forecast data in the future, the number of decomposition modes K and penalty factor α in the VMD algorithm are optimized by the sparrow search algorithm (SSA). The high-frequency signal is assigned to the energy storage system, while the low-frequency signal is assigned to the thermal power units. Second, a double-layer fuzzy controller is designed, in which the fuzzy control in the first layer takes the high-frequency signal of area control error and its changing rate as input to improve the system’s dynamic frequency modulation performance, and the fuzzy control in the second layer takes into account the influence of state-of-charge to prevent the energy storage from encountering problems such as over-charging and over-discharging. Finally, the effectiveness of the proposed control strategy is verified by simulation analysis.

To clarify the internal thermal effect of a battery during its charging process, a lithium-ion battery of 27 A∙h was taken as the research object. A three-dimensional electrochemical-thermal coupling model of one single battery was established, and the coordinate array mapping method was used to expand the thermal effect of the cell unit to that of the single battery. First, the influence of charging rate on the heat production of the battery was investigated. Then, the detailed heat production in the active layer of negative electrode with different particle radiuses was studied, and the mechanism of influence of particle size on the heat production was analyzed. In addition, the battery temperature distribution with different charging rates under natural convection was also studied. The heating mechanism of lithium-ion battery was elaborated upon, and the overall temperature distribution characteristics of the battery was given, providing a theoretical basis for optimizing the battery structure design and developing the battery thermal management system.

To mitigate the fluctuations in the output from micro-sources within a half-bridge converter series Y-connection microgrid (HCSY-MG) grid-connected system and ensure both the balance of the sum of DC-side voltages in each phase and the balance of three-phase grid-connected current, a distributed hybrid energy storage system (HESS) charge-discharge optimal control strategy based on improved proximal policy optimization (PPO) is put forward. With the consideration of grid-connected current in the HCSY-MG system and the characteristics of the distributed HESS, the key system variables affecting the grid-connected current are identified, and the optimal topology for integrating the HESS into the system is determined. Subsequently, by combining the characteristics of the series-connected system, the charge-discharge problem of the distributed HESS is transformed into a Markov decision process within the framework of deep reinforcement learning. Meanwhile, to address the challenge of determining the entropy loss weight in the PPO method, an improved PPO method is proposed, which balances the agent’s convergence and exploration capabilities. Finally, the typical operational data of a renewable energy generation site is used as a case study, and the feasibility and effectiveness of the proposed control strategy is validated.

Self-synchronized voltage source inverters (SSVSIs) have attracted much attention for their capability to improve the voltage and frequency stability of power systems. To solve the problems of energy storage state-of-charge (SOC) imbalance, DC-side voltage fluctuations and AC-side output power oscillations during the parallel operation of PV-storage SSVSIs, the dynamic characteristics of AC-and DC-side power response of a parallel system of PV-storage SSVSIs are analyzed at first. Then, a cooperative control method for the parallel system is proposed, which realizes the dynamic balance of energy storage SOC by considering a cooperative allocation strategy for energy storage SOC and improves the dynamic characteristics of DC-side voltage through the feed-forward control of energy storage power. In addition, the AC-side output power oscillations are suppressed by damping enhancement control. Finally, the effectiveness of the proposed control method was verified by simulations and experimental results based on RT-LAB.

To realize the collaborative control of power mutual assistance, voltage frequency support and state-of-charge (SOC) of energy storage batteries between multiple microgrids, a multi-port interconnected current source converter suitable for the interconnection of microgrids is proposed. Through the introduction of the average SOC of the microgrid group into the active command control loop of a central controller and the addition of a capacity weight factor to the loop, the dynamic equilibrium of energy storage SOCs of three microgrids under various working conditions is realized. In addition, with the consideration of frequency exceeding of the microgrid that may occur during the SOC adjustment process, an SOC-frequency collaborative control strategy that takes into account the frequency performance of the microgrid is put forward, and the frequency operation is guaranteed to be within a specified range on the basis of realizing the SOC equilibrium of the microgrid group. Simulation results verify the correctness and effectiveness of the proposed control strategy.

A two-layer optimization algorithm based on a microgrid load fluctuation suppression strategy is proposed, aiming to address the uncertainties in renewable energy generation and random load fluctuations. Through the modeling of operating states of generation devices and power allocation modes, the upper-layer model mainly considers the micro-resource output power constraints and related constraints on microgrid operations, so as to ensure the scheduling priorities of generation devices while reducing the operating cost of microgrid. The lower-layer model minimizes the equivalent load fluctuations in microgrid while using the values predicted by LSTM-Attention neural network to further constrain the output power of generation devices. The two-layer model is transformed into a single-layer model by applying the KKT condition, thus reducing the complexity in solving the model. Two different load scenarios are designed, and the two-layer optimization algorithm is used to perform simulations under the two scenarios, respectively. Simulation results show that the pro-posed algorithm can dynamically adjust the output power of generation devices and improve the consumption proportion of renewable energy. In addition, it can change the microgrid’s operating state according to different load scenarios and thus mitigate the impact of load fluctuations on microgrid opera-tions.

To solve the problems in a high-permeability regional power grid such as reduced moment of inertia, decreased system frequency regulation capability and lower frequency stability, the frequency regulation control of wind turbines is added on the basis of energy storage frequency regulation. Through the analysis of the influence of wind power permeability (δ) on the system, an adaptive wind-storage coordinated primary frequency regulation control (PFRC) strategy considering the influences of state-of-charge (SOC) and δ is proposed from the perspective of reducing the battery loss and prolonging the energy storage life. This control strategy is realized through the coordination of energy storage adaptive PFRC considering SOC and frequency regulation dead-zone and the wind power adaptive PFRC strategy considering δ, which can effectively improve the system stability and the frequency regulation effect of the power grid. Under the step disturbance, the lowest point of frequency drop changes from 0.33 to 0.07. Under the continuous load disturbance, compared with the single energy storage frequency regulation, the frequency quality under the proposed control strategy is changed from 3.966 to 1.676, and the SOC quality is changed from 0.064 07 to 0.016 48. The data is smaller, and the change in the SOC of energy storage is gentler, which can save the configuration of energy storage capacity and reduce the loss of the energy storage unit. Therefore, a better frequency regulation effect is achieved to maintain the stable operation of the system. Finally, the effect-tiveness of the control strategy is verified by building a regional power grid model in MATLAB/Simulink.

With the significant integration of transportation and energy fields, the electrified and “green” energy consumption in transportation is an important technical solution for achieving the targets of carbon peak and carbon neutrality. The use of low-temperature energy as coolants for high-temperature superconductor (HTS) cables can simultaneously transport the electric energy and fuels, which can improve the energy transmission efficiency and reduce the overall costs for large-scale renewable energy transmission. For a 10 km transmission distance, a 1.5 kV/10 kA liquid hydrogen (LH₂) energy pipeline structure for railway transportation is designed. Meanwhile, the effect of the arrangement angle of positive and negative bipolar HTS tapes on their critical currents was investigated by establishing a finite element simulation model of HTS cable. The maximum critical current was about 15 240 A in the parallel arrangement. Finally, the transmission losses of HTS cables and conventional copper cables with the same capacity (i.e., 15 MW/10 km) were compared. Results show that the transmission loss of 1.5 kV HTS cable was only 15% of that of 10 kV conventional AC cable, and it was also slightly lower than that at the 35 kV voltage level. The findings in this research can provide a feasible technical solution for LH₂ HTS energy pipelines to be used for large-capacity combined hydrogen-electricity transmission.

In power systems, unbalanced and nonlinear loads might lead to voltage and current imbalance and harmonic distortions, which will affect the operation of some critical equipment. To improve the power quality of a system consisting of multiple grid-forming (GFM) converters, it is necessary to reasonably share the load current’s fundamental negative-sequence and harmonic component while reducing the voltage imbalance and harmonics at the point of common coupling (PCC) as much as possible. To solve this problem, a power quality control strategy for GFM converters based on unified unbalanced/harmonic voltage-current droop is proposed in this paper. By establishing a unified droop relationship of fundamental negative-sequence and harmonic component between PCC voltage and output current, unbalanced and harmonic current is shared in accordance with the capacity of each converter while suppressing imbalance and harmonics of PCC voltage. The proposed me-thod can be applied to multiple converters in both the islanded and grid-connected modes without extracting the fundamental negative-sequence or individual harmonic sequences separa-tely. Compared with the existing control methods, this method is simpler and easy to implement in embedded controllers. The design scheme for control parameters based on closed-loop the pole analysis is discussed in detail. Moreover, through a comparison with the existing methods, it is shown that the proposed method has superior dynamic performance and less computation resources. Finally, the effectiveness of the proposed control method was verified by experimental results.

To address the problem of power quality deterioration in a charging pile’s front-stage AC-DC rectifier under frequency fluctuations of public power grid, a current inner-loop control method with frequency adaptation capability is proposed based on the application of a load virtual synchronous machine (LVSM) control strategy. First, a small-signal model of the LVSM frequency loop is established, and analysis results show that the virtual frequency will follow the grid frequency that deviates from the fundamental frequency. Second, a frequency-adaptive quasi-proportional-resonant controller with reduced computational complexity is designed for the current inner-loop to suppress grid-connected current harmonics injection into the grid. Finally, simulation results show that the proposed controller can actively respond to changes in grid voltage and frequency while improving the power quality during the regulation process, which verifies the theoretical analysis.

The grid impedance exerts a serious impact on the stability of grid-tied inverters, and its accurate estimation is conducive to realizing the high-performance adaptive control of grid-tied inverters in weak grid conditions. The PQ variation method is a typical method of online grid impedance estimation. However, research shows that there exists a relationship of nonlinear function between real grid impedances and their estimation results, which may affect the estimation results. The traditional PQ variation method has shortcomings in the grid impedance estimation accuracy due to ignoring the influence of steady-state operating point fluctuations before and after the variation. Therefore, from the two aspects of error compensation with fixed phase-angle difference and estimation value correction with offline look-up table, two improved estimation schemes considering the operating point fluctuations are proposed respectively to achieve an accurate grid impedance estimation under weak grid conditions. The detailed theoretical analysis and test results show that compared with the traditional PQ variation method, the proposed grid impedance estimation method can achieve a higher degree of accuracy.

With the large-scale development and utilization of renewable energy sources such as photovoltaics and wind power, inverters based on droop control are widely used in distributed generation systems and microgrids. Sequential impedance modeling is one of the important methods to assess the stability of a multi-inverter parallel system and optimize the control parameters. However, there is currently limited research on the sequential impedance modeling of droop-controlled inverters. Therefore, a detailed investigation was carried out. First, the steady-state operating point of a droop-controlled inverter was determined, and the inverter response was analyzed by introducing positive-and negative-sequence small signal disturbances. Second, the sequential impedance model was derived from the inverter response to characterize the inverter’s impedance properties. Subsequently, the accuracy of the sequential impedance model was verified through PSCAD simulations. Finally, the effects of key parameters such as droop coefficients and the cutoff frequency of a low-pass filter in the power calculation link on the inverter output impedance were analyzed and verified through experiments. Results demonstrate that the proposed sequential impedance model has a good performance in analyzing the droop control stra-tegies, providing valuable insights for further optimization and practical implementation of the strategies.

With the continuous increase in new energy penetration, the impedance interaction between a grid-connected inverter (GCI) and a weak grid is easy to induce small-signal oscillation instability. To solve this problem, the limitations of the existing voltage feedforward impedance reshaping strategy are analyzed in this paper at first. Then, an adaptive voltage feedforward impedance reshaping strategy is proposed. This strategy utilizes simulated anneal-particle swarm optimization algorithm to optimize the feedforward path parameters off-line under sample conditions of different values of output power and grid impedance, and polynomial fitting is used to cover the continuous full operating range. By deploying the fitting relationship in the inverter, the feedforward path parameters can be adjusted adaptively according to real-time power and grid impedance. The proposed adaptive strategy can ensure a better stability margin of GCI under conditions of different values of output power and grid impedance, and the dynamic performance of the system is considered to some degree. Finally, experiments were carried out on an OPAL-RT semi-physical experimental platform to verify the effectiveness of the proposed strategy.

The modeling and stability analysis methods for a fractional-order Cuk converter closed-loop system are proposed. First, the transfer function of a PWM-controlled fractional-order Cuk converter closed-loop system is analyzed, and the nonlinear links of its control system are modeled to obtain a describing function model for the fractional-order Cuk converter closed-loop system. Then, the effect of fractional-order component order on the stability of the converter closed-loop system is analyzed using an improved Nyquist stability criterion. Finally, the accuracy of the modeling and stability analysis methods for the fractional-order converter closed-loop system was verified by simulations and an RT Box platform.

To address the problems of lower power density and fewer application scenarios of a traditional multi-stage AC-DC bidirectional converter, a novel AC-DC bidirectional converter with adjustable frequency output on AC side is proposed. A matrix frequency conversion circuit is used instead of a full-bridge circuit, and the two working mode control methods of rectification and inversion are combined to realize a single-stage AC-AC frequency conversion function of the converter without DC links, so that the converter can add a flexible energy outgoing function of new energy electric vehicles on the basis of satisfying the vehicle-to-grid energy interaction strategy. At the control level, when the converter works in an inversion mode, proportional resonance control will be used to obtain lower total harmonic distortion and adjustable frequency characteristics, so as to cover more AC-side electrical requirements. When the converter works in a rectification mode, the average current control based on CCM mode can ensure a high value of power factor on the AC side and a stable voltage output on the DC side. Finally, a MATLAB/Simulink simulation model and a hardware experimental platform were built to verify the scheme, and results proved the correctness and effectiveness of the proposed topology.

Five-level inverters are widely used in the fields of energy storage and electric drive systems, especially in the high-voltage and high-power applications. As the power level increases, the capacity of one single inverter unit cannot meet the demand. Therefore, inverters need to be operated in parallel. Compared with the two-level or three-level topology, the circulating current between single-phase modules of parallel five-level inverters is more prominent due to the existence of floating capacitors. An improved current balance open-loop control method for a public DC bus structure is proposed. With a comparison of output current from each power module, a time stamp is set and the time shift control for vector switching is implemented, so that a positive or negative unbalanced current is artificially introduced to quickly eliminate the original circulating current. A set of 6 kV parallel high-voltage inverters were tested, and experimental results validated the effectiveness of the proposed method and the impact of voltage difference of the floating capacitor on the circulating current. The correction process only took 0.1 ms, and the current balance can be quickly and accurately completed.

Compared with the continuous pulse width modulation scheme, the discontinuous pulse width modulation (DPWM) scheme can effectively reduce switching losses and is more suitable for converters with a high switching frequency. However, as the switching frequency increases, the proportion of dead time in the switching period rises, significantly affecting the harmonic performance of the converter. A dead-time compensation method for Si/SiC hybrid three-phase three-level grid-tied converters with DPWM is proposed. When the bridge-leg voltage is high-frequency switching, the impact of dead-time on the switching state will be eliminated by offsetting the modulation wave. When the bridge-leg voltage is clamped, the comparison logic between the modulation wave and the carrier wave will be improved to avoid unnecessary switching actions in the clamped phase and ensure sufficient dead time during the transition of clamping modes. Finally, an experimental platform of a three-phase three-level active clamped converter was set up to verify the feasibility and effectiveness of the proposed dead-time compensation method.

Three-level inverters have attracted widespread attention in medium-and high-voltage applications. However, traditional continuous pulse width modulation (CPWM) methods suffer from drawbacks such as high switching losses, significant common-mode voltage (CMV) fluctuations and neutral-point voltage imbalance. To solve these problems, based on the analysis of the generation mechanism for CMV, the suppression conditions for CMV are derived based on discrete pulse width modulation (DPWM), and a modified DPWM modulation strategy is proposed, which can reduce switching losses under DPWM0~DPWM3 modulations while suppressing CMV. In addition, based on the proposed DPWM modulation strategy, the control method for neutral-point voltage is explored, and the multi-objective optimal control of the three-level inverter is realized ultimately. Simulation and experimental results verified the validity of the proposed DPWM control strategy.

Due to the effects of the droop equation and equivalent impedance, the adoption of traditional droop control as a strategy for parallel-operated inverters will cause steady-state errors in the output voltage amplitude, which is not applicable to scenarios requiring high precision in voltage regulation. To solve this problem, a voltage amplitude compensation strategy based on hierarchical cooperative control for a system of parallel-operated inverters is proposed. The robust droop control is employed as the primary control instead of the traditional resistive droop method, which effectively constrains the amplitude error within a 5% margin and thus ensures the system’s capability to regulate voltage under different operation conditions. On the basis of analyzing the primary control error, the secondary voltage control is carried out utilizing a low-bandwidth communication system, and the voltage amplitude is tuned to the reference value to improve the compensation accuracy. Experimental results verified the feasibility and effectiveness of the proposed scheme.

To minimize the current stress of a dual active bridge (DAB) converter in its full power range, a global optimization strategy for the current stress under triple phase shift (TPS) control is studied. First, the operation principle and process of the converter under TPS control are analyzed, and a model of the relationship among its output power, current stress and phase-shift ratio is established. On this basis, the TPS control is optimized by seeking the optimal phase shift ratio combination with the goal of global minimization of current stress. In addition, the optimization results are compared with those under the dual phase shift (DPS) control in terms of current stress and reactive power. Finally, verification was performed by carrying out an experiment. The theoretical and experimental results show that the TPS control with the global optimization of current stress further reduces the current stress and reactive power in the full power range.

The four-switch Buck-Boost converter applied to a fuel cell system can theoretically achieve a higher efficiency under the traditional two-mode control. However, in fact, the dynamic performance of the system is poor due to the influence of a voltage gain blind zone in the process of transition between the two modes, which greatly reduces the operation stability of the converter. In high-power applications with large current, the interleaving technology is usually used to reduce the inductor current ripple and increase the power density of the converter. Therefore, the smooth mode transition of a four-phase interleaved four-switch Buck-Boost converter is taken as a control objective. First, the topology of the converter is introduced, and its basic mathematical relationship is obtained. Then, the reason for the voltage gain blind zone of the converter is analyzed using a graphical method, and a control strategy is proposed to eliminate the blind zone and realize the smooth mode transition based on the traditional three-mode control strategy. In the dynamic process, the changes in duty cycle are minimized and the inductor ripple current is also optimized, which improves the efficiency and stability of the converter. Finally, the effectiveness of the proposed control strategy was verified by experimental results.