Three-phase bidirectional ac–dc inductive power transfer (IPT) systems for vehicle-to-grid applications are conventionally based on a two-stage architecture. However, poor conversion efficiency, high component counts, and high control complexity are distinct disadvantages. Even though a few single-stage ac–dc IPT systems based on matrix-type converters were proposed, it remains a challenge for achieving sinusoidal input current, zero-voltage switching, and high efficiency simultaneously. In this article, a new single-stage three-phase bidirectional ac–dc IPT converter is proposed. The proposed converter is based on a newly proposed multi-active-bridge converter coupled to a SWISS front end. As a result, the proposed converter inherits not only the merits of low distortion of input current, high efficiency, and output short-circuit protection, but also achieves soft-switching operation resulting from the proposed modulation strategy and the property of the double-sided $LCC$ compensation network. The operating principle and mathematical analysis of the proposed converter are discussed. Experimental results are presented to verify the performance of the converter.

Leakage and magnetizing inductances of a loosely coupled transformer are constantly changing due to the relative positions of the primary and secondary transformer windings in inductive power transfer (IPT) systems. In order to counteract the non-negligible inductances and extract desirable characteristics from the system, various compensation networks have been proposed to obtain unity input power factor and constant output voltage for minimum reactive power and ease of control for battery-charging or other power electronics applications. However, most of the existing control methods with fixed compensation networks, which are usually designed for a specific coupling coefficient, have failed to fulfill both requirements simultaneously due to the mismatch between the resonant frequencies of the networks after the deviation of the designed coupling condition. In this paper, a dynamic series/series-parallel compensation network based on switch-controlled capacitor is proposed to re-match the series and parallel resonant frequencies of the network such that the magnetizing inductance can be adaptively compensated for different sizes of air gap/misalignment. By doing so, the requirements of load-independent output voltage and unity input power factor can be simultaneously fulfilled while zero voltage switching on the primary bridge is still maintained under different coupling conditions by a single-stage converter.

The dynamics of nonresonant dual-active-bridge converter (DABC) are simultaneously affected by the transient modulation strategy and controller design. In general, inappropriate transient modulation strategies can lead to nonzero transient dc offsets in the inductor current and transformer’s magnetizing current, thus introducing excessive trajectory tracking error and time delays between the pulsewidth modulation generator and controller. Consequently, truly optimal transient responses cannot be achieved solely through a high-performance controller, unless the modulation-induced transient dc offsets can be completely eliminated. This article presents a comprehensive review of the optimized transient phase-shift modulation (OTPSM) strategies for single-phase-shift modulated DABC, and derives a novel optimal modulation method referred to as symmetric single-sided OTPSM (SS-OTPSM), which is based on a unified theoretical framework of OTPSM and an additional condition enabling a full elimination of all undesired transient dc offsets. The proposed SS-OTPSM can be easily and cost-effectively implemented in a cycle-by-cycle manner, and inherently compatible with fast controllers. Additionally, in order to more accurately match DABC’s power transfer model under SS-OTPSM, an enhanced model predictive controller (EMPC) is proposed. By a combined use of SS-OTPSM and EMPC, ultrafast and completely dc-offset-free dynamics can be achieved without measuring the inductor current. The effectiveness of the proposed schemes is verified by closed-loop simulation and experimental results.

Linear extended state observer (LESO) plays a critical role in ADRC's control performance. However, there exists a tradeoff between disturbance rejection and noise immunity for traditional linear ESO (TLESO). Compared to TLESO with similar bandwidth, phase-locking loop observer (PLLO) exhibits a better output tracking performance but weaker noise suppression capability. By combining the advantages of TLESO and PLLO, an adaptive LESO (ALESO) is proposed in this article, which utilizes TLESO in steady state and PLLO during transient state. The dynamic and noise-suppression performances of different observers are analyzed and compared systematically. The proposed ALESO inherits good dynamic response of PLLO and noise-suppression performance of TLESO, without using high bandwidth. It also possesses the advantages of simple structure, requiring few tuning parameters and ease of parameter tuning. To verify the feasibility and effectiveness of the proposed ALESO design, it is employed to a 500 W dual-active-bridge dc–dc converter for pulsed power applications. Experiments results show that, in comparison with TLESO, the proposed ALESO has resulted in 33.3%/37% and 60%/73.3% reductions in output voltage undershoot/overshoot and setting time, respectively, under load step-up/down. For steady-state performance, a noise level reduction of 76.3% is achieved compared to that resulting from PLLO.

The increasing penetration of distributed renewable generations has given rise to a novel energy management mechanism, peer-to-peer (P2P) energy trading. The concept of Energy Internet (EI) is proposed as a new energy system framework to facilitate P2P energy trading where all distributed electrical devices are interconnected via energy routers (ERs). A proper market clearing approach is required to coordinate decentralized decision making in P2P market for demand-supply balance and feasible end-to-end energy delivery. In this paper, a decentralized market clearing mechanism considering energy path conflict management is proposed for P2P energy trading in energy local area network (e-LAN). An adjusted conflict-based search (ACBS) algorithm is proposed to deal with the non-convexity of the social welfare maximization problem where the nonconvex problem is transformed into multiple independent convex problems. Considering the privacy protection of participants, a dual decomposition based approach is proposed to solve these convex problems in a decentralized manner where self rationality of each participant can be satisfied by maximizing its personal welfare. In addition, energy path conflicts caused by decentralized optimization is resolved by imposing penalty fee and observing the rule of energy sharing maximization. Numerical simulations are presented to verify the effectiveness of the proposed market clearing approach.

The unique advantages of impedance source inverters make them more suitable for new energy fields that need to boost and invert low dc voltages. In order to enhance the boost capability, this work derives a new high-gain quasi-switched boost inverter (HG-qSBI) from the original qSBI topology. Specifically, the voltage gain is improved from 1/(1-2 d ) to 1/(1-3 d ) by adding one switched-capacitor unit formed by one capacitor, one transistor, and two diodes, here d is the duty ratio for shoot-through states of the inverter. It inherits the advantages of single inductor and continuous input current, and the modulation strategy is exactly the same as the original qSBI. Moreover, all components employed in the HG-qSBI withstand the same voltage stress. All these features are verified by both three-phase simulation and single-phase experimental results.