The back electromotive force harmonics of a five-phase permanent magnet synchronous motor (FPPMSM) will significantly reduce the decoupling performance of a-dual-FPPMSM series drive system powered by a single inverter. Meanwhile, the traditional control strategy for the dual-FPPMSM series drive system based on an accurate motor model is too sensitive to changes in motor parameters. To solve these problems, the deadbeat control of decoupling between torque and flux is constructed at first, with flux amplitude and torque angle as the intermediate variables. Then, the speed ring active disturbance rejection control is established to generate the preset motor fundamental wave torque, based on which the torque ring active disturbance rejection control is constructed to generate the torque angle increment. Experimental results show that the proposed control strategy is insensitive to changes in motor parameters, and the decoupling performance between two motors is better.

The discontinuous pulse width modulation (DPWM) method can effectively reduce the switching loss and improve the system efficiency. However, it will also bring in high-order common-mode voltage, which results in an decline in the quality of grid current and an increase in the common-mode current (CMC). To solve this problem, a novel DPWM method is proposed, which can reduce the content of common-mode voltage in the modulation wave by injecting third-order sinusoidal signals into the wave under the traditional DPWM1 modulation at the clamping stage. A common-mode loop model of a photovoltaic inverter system is established, and the reason of the increasing CMC under the DPWM modulation is theoretically explained. The theoretical analysis and experimental results show that the proposed novel DPWM method can improve the efficiency of the photovoltaic inverter system while ensuring that the THDi and CMC are similar to those under the SVPWM modulation. As a result, good current quality and high efficiency are maintained simultaneously.

The sorting voltage equalization algorithm is often used as a submodule capacitor voltage control strategy for a modular multilevel converter (MMC), which poses a challenge in locating the open-circuit fault of an upper switch in a half-bridge submodule. To solve this problem, a method for locating the insulated gate bipolar transistor (IGBT) open-circuit fault in the submodule of MMC based on reconstructed subunit voltage is proposed. The voltage circuit equation of a three-phase MMC is established, and the influence of current flow path on the capacitor voltage and the reasons for abnormal output current and bridge arm circulating current are analyzed when an open-circuit fault for a single-switch occurs in the half-bridge submodule. An adaptive state observer based on extended Kalman filter (EKF) is used to achieve the optimal estimation of output current and bridge arm circulating current, and fault diagnosis is performed using the residuals between the optimal estimation value and the actual sampling value. Subsequently, the submodules with the same position as the faulty bridge arm in the three-phase system are grouped and reconstructed, so as to obtain the reconstructed subunits. The adaptive state observer based on EKF is utilized to estimate the reconstructed subunit voltage, and the difference between the estimated value and the actual average voltage of the corresponding reconstructed subunit is compared with a preset threshold, thus achieving the fault localization. Four typical IGBT open-circuit fault conditions of submodules were selected for experimental verification, and experimental results show that the proposed fault diagnosis and localization method is correct and feasible.

To improve the reliability of a grid-connected converter under operation, an open-circuit fault diagnosis and fault-tolerant control strategy is proposed for a T-type three-level grid-connected converter. After an open-circuit fault occurs in a power device, through the analysis of the operating state of the converter, the changing relationship between the converter output current and the currents flowing from the midpoint of the DC-side capacitor to each phase is elucidated. The average value of the converter output current and that of the currents flowing from the midpoint of the DC-side capacitor to each phase are calculated, and these values are compared with the preset diagnostic thresholds to locate the faulty power device. According to their locations, the faults are categorized into Class-I (vertical bridge arm fault) and Class-II (horizontal bridge arm fault). The corresponding fault-tolerant structures are designed for the two types of faults, and a fault diagnosis and model predictive fault-tolerant control strategy is proposed for the T-type three-level grid-connected converter. Experimental results verify that the proposed method can quickly locate the position of the faulty power device of the converter, realize fault-tolerant continuous operation and improve the reliability of the converter.

A micro inverter can be directly connected to a photovoltaic board, and it has advantages such as a simple structure, a small size and input output isolation. To expand the power range of the photovoltaic micro inverter, two flyback converters can be interleaved and connected in parallel at the front stage DC-DC section, and an active clamping circuit can be used to improve the efficiency of the micro inverter. In response to the issues of temperature rise in a high-frequency power transformer and low power density in outdoor high-temperature environments, a planar transformer applied to micro inverters is designed. First, the working principle for flyback micro inverters is analyzed. Second, the electrical parameters and winding structure of the planar transformer are designed, and the magnetic density distribution and loss of the planar transformer core are simulated using Maxwell software. Finally, a 220 W interleaved parallel flyback micro inverter prototype was built for experimental testing. Experimental results proved the effectiveness of the design, and the prototype can perform stable work.

In the design of inductors for switching power supply and transformers, the calculation of magnetic circuit parameters is of significance. Given that it is difficult to accurately calculate the core reluctance and core loss of the center-column segmented air gap, an improved magnetic circuit model of the inductor core is proposed with an EC-type inductor core as an example, which is based on the exact domain solution and conformal transformation. Through theoretical analysis, the analytical expression for the equivalent magnetic reluctance of diffused magnetic flux near the center-column air gap in the magnetic core is derived, and a calculation model of magnetic reluctance for the uneven magnetic flux in the inner angle of the magnetic core is put forward as well. Finally, a 3D model was established in the eddy current field of ANSYS Maxwell simulation software, and physical experiments were also carried out. A comparison of theoretical results with simulation results and experimental data verified the accuracy of the theoretical analysis, providing theoretical guidance and reference for the optimization design and industrial production of various magnetic cores of inductors or transformers used in segmented air gaps in switching power supply.

A novel high-gain DC-DC converter based on switched inductor and switched capacitor units is proposed. By configuring one set of switched inductor units and two sets of switched capacitor units, a high gain is achieved while maintaining the low voltage stress and low current stress on each component in the converter. The operation principle for the novel converter in different current modes is explained, and its voltage gain characteristics are derived. In a continous conduction mode (CCM), the current stress on the main components in the converter decreases as the gain increases, so a high efficiency is maintained at high gains. High gains can also be achieved in a discontinuous conduction mode (DCM). In addition, the boundary between the CCM and DCM modes of the converter is derived according to the gain relationship. A 325 W experimental prototype was built to verify the effectiveness of the topology and theory, and it was found that the measured converter gain can reach 18 times and the highest efficiency can reach 94.9%.

To solve the problem of a small output voltage range in single-inductor dual-output Buck-Boost converters, a quadratic unit structure is introduced in this paper, which is the voltage gain by D_i/(1–D_i) times and significantly expand the voltage range. According to the working state of the power supply inductor, the power supply modes of the converter can be divided into four types, i.e., complete inductor supply mode, incomplete inductor supply mode, dual-branch supply mode and single-branch supply mode. The voltage gain, ripple voltage and critical inductance values between each mode of the four power supply modes are calculated. The converter has four output states, i.e., Boost, Buck, Boost-Buck and constant-voltage, and the duty cycle conditions in each output state are derived. Finally, experiments were conducted in four power supply modes and four output states of the converter, proving that the theoretical analysis is correct and the performance of the converter is reliable.

The conventional interleaved parallel Boost converter is widely applied in step-up power supply systems. Although it is theoretically possible to obtain a higher voltage gain by increasing the duty cycle, the voltage gain usually does not exceed 4 times due to the influence of non-ideal factors such as the reverse recovery time of diodes and parasitic parameters. In addition, the gain limit will further decrease with increases in output voltage and power. To solve these problems, a Boost converter with a high voltage gain is proposed to expand the output voltage range at the same input voltage level. Meanwhile, the input and output filter inductors in the converter are integrated into one pair of magnetic cores, so as to eliminate the inherent right half plane zero in conventional Boost converters while improving the power density. Therefore, the output dynamic response speed is significantly improved. The working principle for the converter is analyzed, its characteristics are summarized, and the magnetic core selection method for the coupled inductor is discussed. Finally, the effectiveness of the proposed scheme was verified through simulations and experimental results.

The operating characteristics of DC-DC converters have a profound impact on the flexible scheduling of distributed resources in active distribution network. A modular combined DC-DC converter based on Buck-LLC integration is proposed for new energy resources access. Through the high isolation resonant power unit, the input is connected in series to provide high voltage withstand capacity while the output is connected in parallel to meet the demand of high power, thus realizing the efficient transmission of high power. The proposed topology adjusts the duty cycle of a series dual Buck circuit to meet the requirements of connecting different medium-voltage DC networks. By introducing an auxiliary inductor, the zero-voltage-switching (ZVS) problem of all the switches in a wide input range is solved, and the energy transmission efficiency is improved. Based on a study on the ZVS boundary conditions for each switch, the auxiliary circuit and control parameters areoptimally designed. Finally, a power unit prototype with 6.25–7.00 kW was designed and fabricated, and experimental results verified the feasibility of the proposed topology, as well as the correctness of the aforementioned working principle and steady-state analysis.