Xin Liu

Zero voltage switching (ZVS) control is an effective approach for paralleled inverters with coupled inductors in ac motor drives, as it enhances switching frequency, efficiency, and overall control performance. However, conventional ZVS paralleled inverters only utilize the freedom of zero-sequence injection to adjust switching frequency which faces critical frequency conflicts for different phases, resulting in high current stress and suboptimal power capacity utilization. To address this issue, the freedom of phase-shift operation for paralleled phase-legs is implemented in this paper, and a variable dual phase-shift (VDPS) method under the discontinuous pulse width modulation (DPWM) is proposed for ZVS paralleled inverters. In the proposed VDPS-DPWM method, the circulating current ripple prediction model is proposed with the real time phase-shift angle, allowing the reconstruction of phase-leg current trajectories through the superposition of sampled fundamental current and predicted circulating current ripple. Additionally, an optimized VDPS angle is derived to mitigate critical frequency conflicts. The proposed VDPS-DPWM method demonstrates improved ZVS control and significant current stress optimization across the entire operational range. Simulation and experimental results are provided to validate the effectiveness of the proposed method.

Underwater wireless power transfer (WPT) for autonomous underwater vehicles (AUVs) faces significant challenges, including dynamic misalignment caused by seawater fluctuations, eddy current losses, and wide-range power regulation requirements. To address these challenges, this paper proposes a DC-link parallel AC-link series multi-inverter multi-rectifier (DPAS-MIMR) topology with optimal current control. The proposed system integrates modular topology design with mutual inductance identification, efficiency optimization, and constant-current/constant-voltage charging capability. It can reduce eddy current losses and adapt to seawater fluctuations, which may provide a comprehensive solution for reliable underwater WPT systems. Experimental results show the system can achieve an output power of 2.5 kW with 90.4% efficiency when operating in seawater conditions, validating the feasibility of the proposed system.

The single-stage multi-port inverter (SSMPI) provides a cost-effective solution for multi-source systems thanks to its compact design that eliminates the need for additional dc/dc converters. However, the modulation scheme for SSMPI is challenging to design, because the degrees of freedom (DoFs) provided by conventional carrier-based or space-vector frameworks are insufficient for synthesizing ac voltage and distributing power among all dc ports simultaneously, especially when the number of ports is large. To bridge this research gap, this paper proposes a general algebraic modulation scheme that provides sufficient DoFs for simultaneously achieving ac voltage synthesis and power distribution, regardless of the number of ports. In the proposed scheme, the modulation design is formulated as a straightforward solution process for algebraic equations from a switching-duty-cycle perspective, allowing the available DoFs to expand autonomously as the variable space increases. Based on this concept, the modulation implementation is further carried out using a two-step method, where the duty cycle differences are first solved to guarantee power distribution performance, and the volt-second balance constraint is then calculated for ac voltage synthesis, providing the advantages of easy implementation and low computational burden. Finally, the effectiveness and advantages of the proposed scheme are verified through simulation and experimental results.

The active power decoupling technique is a promising solution for grid-connected converters under unbalanced grid conditions due to its advanced merits in simultaneously achieving negative-sequence current elimination and active power ripple suppression. However, an extra power decoupling circuit, along with a complex power ripple tracking controller, should be designed for these purposes. Given this, this article proposes an autonomous power ripple suppression scheme that achieves both active power ripple suppression and negative-sequence current elimination based on a grid-connected multiport converter. In the proposed scheme, negative-sequence current elimination is implemented at the ac port, and the dc-port power is controlled to be constant to suppress active power ripples. Thus, the differential power ripples are autonomously buffered by the ripple port. Besides, the multiport converter is decomposed into an upper-port converter and a lower-port converter to ease the modulation design under time-varying dc-link voltages. With the proposed scheme, balanced three-phase grid currents and ripple-free dc output voltage are simultaneously obtained under unbalanced grid conditions without extra power ripple controllers. Experimental results are presented to demonstrate the effectiveness of the proposed scheme.

The multisource inverter (MSI) serves as a high-efficiency interface for hybrid energy electric vehicles by eliminating the need for intermediate dc–dc converters. Nevertheless, modulation design for the MSI is challenging as the motor drive on the ac side and power distribution between the energy sources on the dc side should be achieved concurrently. Besides, unbalanced dc port voltage of the MSI stemming from the direct energy storage integration leads to a heavy computational burden under the asymmetrical space vector diagram. Given this, this article introduces a modulation waveform split (MWS)-based power flow control for MSIs, aiming to achieve power flow control on the dc side and motor drive control on the ac side concurrently with easy implementation. In the proposed scheme, a carrier-based pulse-width modulation with third-order harmonic injection is utilized to address motor drive control on the ac side, and an MWS scheme under the virtual two-level inverter is proposed to achieve power flow control between the dc sources. With the proposed scheme, control objectives on both ac and dc sides can be met simultaneously in the carrier-based modulation framework, without the complexity of voltage vector synthesis under an asymmetrical space vector diagram. Besides, the power conversion efficiency can also be improved due to the discontinuous characteristics of the split-obtained duty cycle waveform. The steady-state and dynamic performances validate the effectiveness of the proposed method.

The single-stage multiport inverter (SSMPI) is a high-efficiency and cost-effective solution for islanded microgrids with hybrid energy source integration owing to the elimination of intermediate dc–dc converters. However, achieving simultaneous ac voltage control and dc port power distribution remains challenging due to inherent nonlinearities in power distribution and coupling between control objectives. Given this, a proximal policy optimization (PPO)-based reinforcement learning (RL) control scheme is proposed to achieve both ac voltage control and power distribution in SSMPI. In the proposed method, nonlinear features and coupling issues of the SSMPI are formulated as a Markov decision process (MDP) comprising the state, the action, and a reward function that encodes the constraints and objective priorities. A deep actor-critic network is designed to efficiently capture the nonlinear mapping between the environment and control variables. Based on this formulation, offline training enables the agent to learn the optimal policy, while sensitivity analysis of reward weights ensures stable and balanced control performance. Experiment results demonstrate that the proposed scheme effectively controls ac voltage and dc power despite nonlinearities and coupling, reducing power ripples by 52.22% and 88.97%, and ac voltage THD by 27.34% and 83.44% on average over previous methods.