Simultaneous non-contact energy transmission for multiple loads with separate receiving coils is a remarkable advantage of magnetically coupled resonant (MCR) wireless power transfer (WPT). A novel multi-frequency driver configuration for MCR WPT system with multiple loads is proposed, where the receiving resonant tanks are severally tuned at different resonant frequencies to make the loads selective. A methodology of adjusting the resonant frequency of the transmitter is presented to satisfy diverse energy requirements of specific loads. However, the cross interference from non-targeted frequencies inherently exerts an influence between multiple receivers, and thus makes power allocation non-constrained through multi-coupling channels, which increase the difficulty of power control and parameters design. In order to reduce and further eliminate the influences of the cross interference and realise targeted power distribution for selective loads, three types of parallel and series resonant compensation circuits are introduced into receivers. By flexibly utilising the impedance characteristics of parallel and series resonant networks, the proposed compensation circuits are proved to be effective, significantly reducing the cross interference and exclusively delivering power to the selective loads corresponding to pre-tuned resonant frequencies. Finally, the experimental results from a prototype have confirmed the effectiveness of the proposed methods.

The need for emulation of electrical load is often experienced in industrial design, development, and testing of electrical sources, especially in the case of renewable energy sources. The performance of standalone renewable energy systems is found to vary considerably with a change in load characteristics, especially with varying power factor (PF) and harmonics. Here a novel load emulation technique has been developed, using which variable PF and variable harmonic loads can be easily emulated and the energy drawn by the load can be recycled. In the proposed technique, the current drawn by the load is divided into in-phase, leading and lagging components. Each component is controlled individually. The in-phase power is then regenerated. The simulation and hardware results show exact variation in load characteristics as per the test requirements, which are difficult or impossible using passive loading and existing load emulation techniques with back to back converters. The proposed method is found to be more accurate, flexible and easier for emulation of low PF and high harmonic loads. A performance study is also conducted on a solar photovoltaic system using the developed model, substantiating the importance of this load emulation method.

Two medium voltage direct current (MVDC) grids can be connected via an isolated DC–DC modular multilevel converter (DC–DC MMC) to enhance the power transmission and supporting abilities. DC power, DC terminal currents, arm energy balancing and active/reactive power through inner AC link need to be controlled in such a DC–DC MMC, and the submodule and grid fault situations should also be addressed. Therefore, a multi-object controller is inherently required. This study thus proposes a multi-object control frame that can realise not only DC power regulation but also inner dynamic control, reactive power optimisation and fault riding through. To facilitate the controller design, a detailed dynamic model of DC–DC MMC is also derived, and such a model is turned into different forms according to different control objects. The proposed control method and the main design considerations are given in a detailed mathematical way. A full-scale simulation system containing a 6 MW DC–DC MMC, a 20 kV MVDC grid, a 10 kV MVDC grid, and several sources and loads is built to verify the control compatibility, and a down-scale low-power prototype is also built to test the control performance.

Neutral shifting (NS) is a popular scheme to achieve fault tolerance operation (FTO) of multilevel inverters (MLIs) such as cascaded H-bridge (CHB). This fault tolerance scheme (FTS) can be realised with/without zero-sequence voltage injection. Among these, NS with zero-sequence injection FTS is relatively easier to implement. However, this scheme is not generalised for multiple open-circuit faults. Moreover, NS-FTS schemes are not directly applicable for fault tolerance of reduced switch count (RSC)-MLIs, as these inverters have limited redundancies. Therefore in this study, FTO for RSC-based multilevel dc-link (MLDCL) inverter using NS zero-sequence injection FTS is proposed for simultaneous failure of multiple switch faults. Generalised mathematical equations are derived to calculate the magnitude and phase angle of injected zero-sequence voltage for obtaining balanced line voltages with uniform power sharing among all healthy units. The proposed generalised NS-FTS with zero-sequence injection is implemented on three-phase 15-level MLDCL inverter for various fault conditions. The obtained simulation results are validated experimentally on nine-level MLDCL inverter.

Here, a novel control strategy combined by constant-frequency V ² control and H _∞ robust control is proposed and tested in a buck converter. The control strategy not only demonstrates the fast transient response of ordinary V ² control, but also shows strong robustness with different input conditions. The main drawbacks of V ² control, such as high interference susceptibility and low input perturbation stability margin, can be improved by introducing H _∞ robust strategy into the controller design. During the analysis and design process, the original controlled system based on the V ² control is modelled at first, then the H _∞ weighting functions are calculated according to the small-signal analysis of V ² controlled buck converter. After that, the H _∞ robust compensator parameters based on constant-frequency V ² control are calculated. Finally, a buck converter controlled by the proposed control strategy is built and tested, with 48 V input voltage, 12 V output voltage, and 120 W output power. The experimental results demonstrate the effectiveness of the converter with the proposed control, which also proves the proposed method has great potential for high robustness and fast dynamic response applications.

The authors propose a novel step-up converter with stackable switching stages that is suitable for renewable energy applications. On the one hand, the converter gain corresponds to that of the traditional quadratic boost converter, achieving an arbitrary exponential gain in extended configurations. On the other hand, the proposed converter requires a single switch, while the output voltage is partitioned among several capacitors. As argued in this work, the features of the proposed topology represent a significant contribution with respect to standard topologies that exhibit greater voltage stress. The operation principle and the main characteristics of the proposed converter are validated with experimental results.

A new high-efficiency resonant boost converter is presented. The topology is a hybrid circuit between switched-resonator and switched-capacitor converters. The proposed converter can efficiently provide voltage conversion ratio between 1 and 2. The voltage gain is adjustable, and full-range zero-current switching operation is provided for switches independent from operating voltages and load power. The structure is formed by an inverter arm without employing any additional switch in the circuit. Passive elements are all high-frequency resonant type and then small size realisation is possible. The converter stability is guaranteed. Complete theoretical analysis along with simulation and practical results are presented. Finally, the converter is compared with its most recent counterparts.

In this study, a current sensorless sliding mode control (SMC) approach is proposed for a direct current–alternating current inverter system. The SMC scheme shows advantages in dealing with matched disturbances and model uncertainties. However, it fails to remove the effects of mismatched disturbances. In distributed generation systems, wide range of load variations and effects of non-linear loads result in performance degradation. Load disturbances, as the mismatched disturbances, entry the system via a channel different from the control input. Load variations are coupled with the output voltage, resulting in periodic disturbances. Meanwhile, the cost increases significantly due to the requirement of wideband current sensors. In the existing current sensorless approaches, observers are designed to reconstruct current without considering disturbances. Addressing it, a novel universal state and disturbance observer (USDO) is developed to reconstruct the inductor current and load disturbances. By integrating the estimations into the design of sliding surface, a current sensorless sliding mode controller is proposed. Rigorous stability analysis for the closed-loop system is presented. Furthermore, considering higher-order harmonics, the baseline USDO is extended to reject these harmonics. Experimental results are illustrated to verify the effectiveness of the proposed scheme.

Realisation of high-efficiency modular multilevel converters (MMCs) during wide AC grid voltage variation is a significant challenge for the voltage-source-converter based high-voltage direct-current system. In this study, the DC component of the sub-module output voltage is investigated to be a new control freedom by building the inherent relationship between the DC voltage component and active sub-module number. Then, the effect of AC voltage on DC voltage component is explored. The adaptive DC voltage component modulation is proposed to dynamically regulate the active sub-module numbers during wide AC voltage variation, resulting in the improved MMC conversion efficiency. Moreover, the detailed performance analysis shows that the adoption of the proposed modulation brings little effect on the sub-module capacitor voltage/current ripple, circulating current suppression as well as the AC and DC terminal voltages. Finally, the simulation and experiment results show that the adaptive DC voltage component modulation is effective to reduce the power losses during a wide AC voltage range.

This study presents a new topology of switched-capacitor (SC) multilevel inverter, which is able to step-up input DC voltage to a multilevel AC waveform. This single source inverter is designed based on series connection of the capacitors that charged by input DC sources through a SC network. The proposed modular inverter uses famous T and cross-connected modules that can be simply extended to higher output voltages without increasing the amount of total standing voltage and peak inverse voltage of switches. It generates positive and negative voltage levels inherently, which eliminates requirements of H-bridge inverters that are traditionally used to achieve a bipolar output voltage. Analysis shows that the voltage stress on components, cost, efficiency and losses are kept in acceptable range especially for higher-voltage levels. Capacitor's voltage self-balancing is another inherent advantage of this modular topology which leads to simplify control strategy and eliminate excess balancing circuit. Performance of a six-step proposed structure is evaluated by theoretical analysis, simulation and experimental results.

Multilevel inverters (MLIs) have developed deep roots in various industrial sectors owing to their advantages over conventional two-level inverters. However, the reliability of the semiconductor devices has been one of the major concerns for the proper functioning of MLI. Therefore, a novel fault-tolerant topology is proposed in this study. The proposed topology is capable to tolerate single- and multi-switch faults. It has lesser device count compared with the most recent work in the field. Moreover, it achieves inherent voltage balancing across capacitors. The proposed fault-tolerant topology is simulated in MATLAB/Simulink and validated experimentally.

The three-level diode-clamped matrix converter (TLDCMC) topology has outstanding performance under ideal operating conditions. However, input disturbance can influence the waveforms at the output side of the converter due to the direct coupling between the input and output. This study proposes a modified modulation scheme for TLDCMC during operation with unbalanced input voltages and when different transformer turns ratios are used for an isolation transformer at the input. With this modulation technique, sinusoidal and balanced output voltages are guaranteed and the input current harmonics are minimised. Experimental results are presented to demonstrate the feasibility and effectiveness of the proposed modulation scheme.

As the grid inductor increases the phase difference between the inverter admittance and the grid admittance increases at their intersection frequency showing that the harmonic amplification coefficient is enhanced. When the phase difference exceeds 180° the admittance ratio cannot satisfy Nyquist criterion and the system becomes unstable. To address this problem a phase-reshaping strategy aiming to revise the phase of the inverter admittance in a wide frequency region without using any real-time information referring to the grid impedance or the resonant frequency is proposed. The strategy is realised by the combination of the feedforward of the voltage at the point of common coupling (PCC) and the harmonic compensators. In the feedforward, a phase lag network is employed to modify the admittance phase in a wide frequency region. The harmonic compensator in parallel to the current controller is used to offset the magnitude rise brought in by the PCC voltage feedforward. After adopting the proposed strategy, the upper limit of the phase difference is constrained at 150° and the magnitude of the inverter admittance is lower than that of the original one at low-order odd-harmonic frequency. The validity of the proposed strategy is verified by the experimental results.

Large electrolytic capacitors are normally applied to maintain a stiff DC-bus in uninterrupted power supply systems. However, the low-frequency currents flow through them, which can reduce their lifespan and risk the system reliability. Therefore, this study investigates the neutral-point current and corresponding suppression scenarios for three-phase four-wire three-level T-type inverters. First, the neutral-point current for three-level T-type inverters is analysed and the mathematical expression is obtained. With the mathematical model, the neutral-point currents in cases of different load conditions are investigated. In order to reduce the neutral-point current and extend the lifespan of DC-bus capacitors, a neutral-point current suppression control strategy is proposed. The basic concept of the proposed control strategy and its effectiveness in cases of different load conditions are presented. Finally, a 30 kW T-type three-level inverter platform is built and the experimental results are presented to verify the theoretical analysis.

A buck-boost full-bridge topology is preferred for low-input voltage, high gain, high-power battery fed front-end converter of an all-electric motor drive system. Variants of this topology feature high direct current (DC)-gain, soft switching, continuous input and output currents with a phase modulation/asymmetrical pulse width modulation (PWM) scheme. These variants exhibit a high-magnitude DC-current in the transformer primary winding for low-input voltage, high-power application leading to poor core utilisation and low-power density of the system. The operation and analysis of the converter are presented to highlight the DC-current in the transformer and to mitigate this, a hybrid control scheme (HCS) is proposed. The proposed HCS consists of a DC-current compensation (DCCC) loop to mitigate the DC-current in the transformer without altering the DC-gain with no additional components and output regulation (REG) loop to regulate the output. The output REG and DCCC loops are independent of each other. Soft switching is retained with this proposed scheme. The necessity to mitigate the DC-current, analysis, and implementation of the HCS is discussed. The verification of the proposed HCS scheme in simulation and the experimental prototype is presented.

This study proposed a high-gain dc–dc converter with an interleaved current at the input stage and voltage series connection at the output stage. This converter was designed based on the boost–flyback topology, which has as main characteristic high voltage gain. Normally, the boost–flyback is designed with the low input voltage, and therefore, the current at this stage reaches high current values, limiting the converter design. Therefore, paralleling the input connection allows the current division between the semiconductors, reducing the switching stress. As part of the converter analysis, in this study, the static and the dynamic converter model are obtained to further design the converter controllers. Furthermore, simulations and experimental results for the proposed converter, operating in closed-looping control, were designed for a peak power of 1.2 kW, reaching a maximum efficiency of 97.5%.

Multilevel inverters (MLIs) are nowadays extensively used in integration with renewable energy sources and in drives applications. The converter cost and voltage quality improvement are the utmost importance in such application of MLI. A six-switch seven-level reduced switch symmetrical MLI (6S-7L MLI) is proposed here. Compared to the conventional and similar existing MLI topologies, the proposed MLI uses less active switches and has less driver circuit requirement. A particle swarm optimisation (PSO)-based modified selected harmonic elimination technique is derived and analysed for computing optimal switching angles of proposed 6S-7L MLI to eliminate third- and fifth-order harmonics. Moreover, the performance of the proposed algorithm is compared with the two most commonly used PSO variants. The analysis shows the modified version of PSO is most suitable for optimising the output voltage of proposed MLI through targeted harmonic elimination. The proposed topology is investigated through simulation by applying the calculated switching angles using modified PSO algorithm. Finally, a single-phase experimental prototype is designed to verify the validity of the proposed structure.

Inductive power transfer (IPT) has attracted a lot of attention in recent 30 years due to its advantages of convenience, safety, reliability and weather proof. This study proposes a modified parameter tuning method for primary inductor-capacitor-inductor, secondary parallel (LCL/P) compensation topology, which provides the characteristic of excellent constant current output (CCO). This characteristic is analysed on the basis of inductor-capacitor(LC) and capacitor-inductor (CL) resonant tanks. To lower the voltage stresses over capacitors and current stresses through inductors, the stress-based optimisation is conducted. The sensitivity of CCO characteristic with respect to the normalised deviation of compensation parameters is analysed; therefore, the performance of LCL/P compensated IPT system in practical scenarios can be fully understood. The loosely coupled transformer (LCT) is specially designed since it has a significant impact on system efficiency, component stress, power density etc. An exhaustive search algorithm is employed to find the optimal dimensions of the LCT. A 100 W LCL/P compensated IPT prototype is fabricated. The measured end-to-end efficiency, from DC input to DC output, is as high as 92.8%. The output current only increases by 2.02% when the load decreases by half. All experiment results agree well with the theoretical analysis.

This study presents finite control set model predictive control (FCS-MPC) methods to eliminate leakage current for a three-level T-type transformerless photovoltaic (PV) inverter without any modification on topology or any hardware changes. The proposed FCS-MPC methods are capable of eliminating the leakage current in the transformerless PV system by applying the defined candidate voltage vector (VV) combinations with only six medium and one zero VVs (6MV1Z) or three large and three small VVs, which generate constant common-mode voltage to perform the optimisation in every control period. With fewer VVs used for the optimisation, the computational burden can be significantly reduced. Furthermore, comparative analysis is performed to show that among these proposed methods, the 6MV1Z method can achieve satisfactory performances in both grid current tracking and neutral point potentials balance control even with less number of candidate VVs, which exhibits the FCS-MPC as an alternative control strategy to be used in the grid-connected transformerless PV system. Finally, experiments are performed to validate the analysis and the effectiveness of the proposed methods.

This study presents a control method for isolated bidirectional dual-active-bridge converters to optimise the efficiency by achieving the zero-voltage switching (ZVS) in high-voltage applications, which the switching losses are the major issues. At light-load condition, the duty control algorithm is proposed to achieve the ZVS for all switches while the triple phase shift with ZVS condition is realised at heavy load condition. The operating principles and the design considerations of the proposed control scheme are analysed in detail. A 2 kW prototype converter is built up and tested to verify the feasibility of the proposed control strategy.