Journal of Power Electronics最新文献

This paper proposes a smooth mode-switching method based on state trajectory control to suppress overshoot and to shorten switching time during the mode switching of LLC-C resonant converters. First, the resonant tank trajectories of the LLC and LCCL are analyzed. Second, through a transformation of the resonant tank trajectory, the optimal trajectory of the resonant tank switching is drawn. Then the switching optimization cycle time is calculated by a diagram of the optimal trajectory. Thus, the PWM conversion to the optimization cycle is controlled directly when the switching signal comes. By this control method, a smooth transition of the resonant tank voltage and current between the two modes is achieved. At last, a prototype with a rated power of 500 W is built to check the feasibility and effectiveness of the proposed switching method. Experiment results show that the current surge of the resonant tank is reduced from 12.3 to 8.6 A when the state trajectory control is applied. The voltage surge of the second resonant capacitor in parallel is reduced from 906 to 712 V. Meanwhile, the switching time is shortened by 0.21 ms, which speeds up the switching process.

The mechanism of a ZVT PWM boost converter is analyzed in detail in this study. In addition, the comprehensive parasitic parameters are taken into consideration. It is discovered that some unexpected modes, referred to as sneak circuit modes, occur during the operation of the converter, when certain excitation conditions are applied. Parasitic parameters and dynamic sneak paths impact the converter operating performance, causing both positive and negative effects. The effects of sneak circuits and excitation conditions are analyzed so that those with negative effects can be eliminated, and those with positive effects can be utilized. In contrast to previous works, sneak circuit phenomena are utilized to optimize the converter performance. Furthermore, parasitic circuit elements are identified and quantitatively determined to estimate their impact on the switching performance of the converter. Combining parameter optimization and topology optimization, soft switching is achieved in both the main power switch and the auxiliary power switch, and waveform distortion is reduced. Finally, simulation and experimental results are included to verify the theoretical analyses.

Planar magnetic components are compact and less susceptible to skin effect due to their thin copper layers. However, the increase in parasitic capacitance has been a challenge in high-frequency operation of planar transformers (PTs). Parasitic capacitances resonate with inductance and may damage switch devices. Thicker printed circuit board (PCB) and reconfigured windings are suggested in this paper to mitigate the parasitic capacitances. A detailed analysis of the parasitic capacitances is performed with a prototype PT. The resonant frequency increased from 1.27 to 1.63 MHz with a 0.4 mm thicker PCB. The reconfigured PT was 1.38 MHz, 0.3 MHz higher than the original PT.

The core equipment of lithium-ion battery energy storage stations is containers composed of thousands of batteries in series and parallel. Accurately estimating the state of charge (SOC) of batteries is of great significance for improving battery utilization and ensuring system operation safety. This article establishes a 2-RC battery model. First, the Extended Kalman Filter (EKF) algorithm is used to obtain preliminary SOC estimates. Then, the updated error values of the Kalman matrix, the state variables obtained from the EKF algorithm, and the battery data during system operation are used as the training and test dataset for the Long Short-Term Memory (LSTM) neural network algorithm. Finally, the algorithm was compared and analyzed with commonly used EKF estimation methods and LSTM algorithms. It was found that the root-mean-square error of the SOC of the EKF–LSTM algorithm under different operating conditions was less than 0.8%, and the average absolute error was less than 0.5%. The estimation accuracy is higher than either the EKF algorithm or LSTM algorithm alone.

The main methods for suppressing the common mode (CM) noise in flyback converters include adding a shielding layer and adding a balancing winding. To combine the advantages of both methods, a balancing winding with a shielding effect is proposed to suppress CM noise. First, the CM noise transmission path of a flyback converter is analyzed, and an equivalent model of CM noise is established. Then, combined with the CM noise characteristics of the flyback converter, the design criteria for the shielding-balancing winding are given. Finally, experiments are conducted on a flyback converter prototype. The CM noise spectra of four transformer structures (without a shielding layer, with a traditional copper foil shielding layer added, with a balancing winding and with the proposed shielding-balancing winding) are compared. The results show that the CM noise of the flyback converter can be effectively suppressed by adding the shielding-balancing winding proposed in this paper. In addition, the effectiveness of the proposed method is verified.

When a series arc fault (SAF) occurs in one branch of a low-voltage alternating current power distribution system with complex connections and many types of loads, the load branches interact with one another, and thus, detection becomes more difficult. To avoid the occurrence of electrical fire, an SAF detection method based on a residual deformable convolutional network (RDCN) is proposed. First, an arc fault data acquisition experimental platform for low-voltage complex branches is built to measure trunk current signals during normal operation and SAF occurrence. Second, 1D current signals are mapped to 2D space as input to the model, more comprehensively responding to the amplitude and variation of the signals. Deformable convolutional networks are used to extract spatial distribution information from feature maps to avoid the limitation posed by the fixed shape of convolutional kernels and improve spatial differentiation among different data. Considering the ability to focus better on fault information, channel attention is introduced to strengthen the correlation among feature channels. Then, the experimental platform data verify that the method can effectively distinguish whether SAF occurs and the type of load where the fault occurs, with the highest accuracy of up to 98.98% and 98.84%, respectively, and an average test time of 1.8 s in determining whether a fault occurs in a six-branch circuit. Finally, compared with existing network models, RDCN has a shorter running time while obtaining a higher accuracy rate.

In the operational context of the hybrid modular multilevel converters (MMC), the submodule stands out as a pivotal and vulnerable element, susceptible to various faults (Zheng et al. in IEEE Access 7:34946–34953, 2019; Xue et al. in IEEE Trans Power Deliv 33:1448–1458, 2017; Lee et al. Reactive power control operation scheme of LCC-HVDC for maximizing shunt capacitor size in AC systems. IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC). pp 1–5, 2015; Lin et al. in IEEE Trans Power Deliv 31:1342–1350, 2015; Debnath et al. in IEEE Trans Power Electron 30:37–53, 2014), including insulated gate bipolar transistors (IGBT) open-circuit, short-circuit, capacitor faults, etc. Therefore, accurate and rapid diagnosis of submodule faults is crucial to maintaining the reliability of both the hybrid MMC and its associated DC system (Xu et al. in IEEE Trans Power Electron 31:2720–2729, 2015; Wang et al. in IEEE Trans Power Deliv 32:1535–1544, 2016; Guo et al. in IEEE Trans Power Deliv 32:666–677, 2016; Zeng et al. in IEEE Trans Power Deliv 30:1298–1306, 2014). To enhance the reliability of the hybrid MMC, a comprehensive study on fault analysis and a diagnosis scheme based on waveform similarity for the submodules in hybrid the MMC are proposed. The devised scheme leverages a sliding window approach to capture essential feature data and employs partial Hausdorff (P-Hausdorff) distance for the diagnosis of open-circuit faults within the submodules. Finally, the proposed diagnosis scheme is implemented in MATLAB and the effectiveness and validity of the scheme are verified.

Severe torque ripple and determination of the weighting factor pose significant challenges in traditional predictive torque control. A deadbeat predictive torque control scheme designed to minimize torque ripple for dual three-phase permanent magnet synchronous motors is proposed in this paper, eliminating the need for a cost function. Additionally, a novel method for calculating action time based on geometric principles is proposed. First, the deadbeat direct torque and flux control is employed to obtain the reference voltage vector. To reduce the computational burden, the nearest virtual voltage vector is selected directly without a cost function. Subsequently, the torque ripple is further reduced by calculating the corresponding action time based on the introduced geometric relationship in the volt-sec coordinate system. Finally, experimental results demonstrate the effectiveness of the torque ripple reduction method under different working conditions.

Taking the photovoltaic-battery dual-input LLC resonant converter as the focal point of research, a mode-switching switch is incorporated to address the narrow range of voltage regulation in frequency conversion control. A wide-gain composite control strategy, with combined frequency conversion control and variable-mode control, is proposed. The variable mode control facilitates switching between the different operational modes of the converter, allowing for the regulation of the output power of the load while expanding the voltage regulation range. However, variable mode control can result in significant ripple in the output voltage. As an auxiliary voltage regulation method, variable frequency control is used with variable mode control to reduce ripple. Based on the wide gain composite control strategy, the implementation conditions of soft switching in wide gain LLC resonant converters are analyzed in detail to improve converter efficiency. Finally, the feasibility and effectiveness of the proposed composite control strategy are validated through rigorous experiments.

Dual-output three-level converters are widely used in multiphase motor drives, and their common-mode voltage (CMV) and neutral point (NP) balance problems can affect the safe and stable operation of systems. Owing to their unique structure, certain disabled switching states exist. Suppressing the CMV to comply with operational constraints is worthy of further investigation. In this paper, an improved time-sharing virtual space vector (ITVSV) modulation strategy is proposed. In addition, the working principle of the converter is analyzed and divided into two effective working modes. The proposed strategy defines two kinds of virtual vector synthesis methods, using virtual vectors instead of the basic small vector of large CMV to reduce the degrees of freedom of voltage fluctuation. The switching loss can be reduced by adjusting the switching sequence in some areas. Furthermore, it is demonstrated that the NP voltage is self-balancing within one primitive period. Finally, the effectiveness of the ITVSV for suppressing CMV and NP voltage balance is verified by experimental results.