Model predictive control is a promising approach to control a multi-objective modular multilevel converter (MMC). In this approach, the control objectives of MMC are included in a single cost function and evaluated for all possible switching states by using weighting factors. The complexity of weighting factor selection process increases with the number of control objectives, and it affects the performance of MPC as well. To reduce the dependency on the weighting factors, a new model predictive control with zero-sequence voltage injection is proposed. With the proposed approach, some of the control objectives like reduction of submodule capacitor voltage ripple and circulating current can be achieved without using a cost function. The proposed approach also reduces the total harmonic distortion of the output voltage and current waveforms. The performance comparison of the proposed and existing MPC approach has been verified through MATLAB simulations on a three-level flying capacitor (3L-FC) based MMC system.

Multilevel current source converter (MCSC) is an effective way to increase the power rating of conventional current source-fed motor drive, especially its current rating. However, there are still several challenges existing in this kind of configuration, which mainly focus on its space vector modulation (SVM) technique and the inherent DC current imbalance issue. In this paper, an SVM scheme with optimal vector sequence and DC current balancing capability is proposed for five-level current source converter (5L-CSC). A kind of five-segment sequence is finally determined as the optimal one over high modulation index operating range. Moreover, a DC current balancing method, which is involved into the SVM scheme, is also designed in this paper, in which the switching states, which generate the small vectors, are selected based on the magnitudes of the DC current differences at positive and negative DC trails. The proposed SVM scheme is verified on a (2MW/4160V/278A) back-to-back 5L-CSC system through simulation.

Multilevel current source converter (MCSC) is an effective way to increase the power rating of conventional current source-fed motor drive, especially its current rating. However, there are still several challenges existing in this configuration, which mainly focus on its modulation and control techniques, and the inherent DC current imbalance issue. In this paper, a model predictive control (MPC) scheme with DC current balancing capability is proposed for a five-level current source converter (5L-CSC). The optimal space vectors for 5L-current source rectifier (5L-CSR) and current source inverter (CSI) are selected for their respective control objects. Moreover, the DC-side model of a 5L-CSC is analyzed, which is used for DC current prediction based on the applied switching states at both 5L-CSR and CSI sides. Finally, optimal switching states are selected and applied to realize DC current balancing. The proposed MPC scheme is verified on a (2MW/4160V/278A) back-to-back 5L-CSC system through simulation.

Predictive control has recently emerged as a promising alternative to more traditional methods for the control and modulation of power converters. This paper presents an overview of predictive control techniques applied to matrix converters. The paper highlights that predictive control strategy is a promising alternative to conventional modulator based linear control for matrix converters due to its simplicity and flexibility to include additional constraints within the control to have it suitable for different applications. In addition to describing many advantages of predictive control techniques, its limitations and weaknesses are also discussed along with some future trends and applications. Most important control aspects of predictive control are demonstrated through simulation analysis.

The multilevel submodules (SMs) bring the added advantages of compact size and improved efficiency over standard submodules to the modular multilevel converter (MMC). These SMs have more than one floating capacitor, which stores the arm energy. Each arm of MMC has several SMs connected in series to meet the required operating voltage. The SM capacitors voltage or energy must be controlled to ensure the safe and reliable operation of MMC. This objective can be achieved by using either voltage balancing or energy balancing approaches. When all the SMs have equal capacitance value, then the voltage and energy balancing approaches have identical performance. However, the SM capacitance deteriorates over a long run, which affects the performance of MMC. This paper presents the performance comparison of voltage and energy balancing approaches for MMC with equal and unequal SM capacitance. The performance of balancing approaches is evaluated in terms of active and reactive power loading of individual SMs, rating of IGBT devices and SM capacitors, and output voltage and current harmonic distortion.

The computational complexity is one of the major issues in the implementation of direct model predictive control (D-MPC) for modular multilevel converters (MMC). The computational complexity is significantly reduced by combining the conventional voltage balancing approach with the MPC. This approach is referred as indirect model predictive control (I-MPC). The D-MPC and I-MPC approaches have higher voltage harmonic distortion and current ripple. In this paper, an improved indirect model predictive control (II-MPC) is proposed to overcome the drawbacks of D-MPC and I-MPC approaches. To implement the II-MPC strategy, a detailed three-phase MMC model is presented. The simulation studies are conducted on a three-phase half-bridge based MMC to validate the dynamic and steady-state performance of II-MPC strategy.

The use of conventional half-bridge submodules in the Modular Multilevel Converters brings important benefits such as the reduction in the power losses and semiconductor devices. However, the system with half-bridge submodule lacks the capability of blocking DC-faults currents. In this paper, a 5-level submodule with DC-fault blocking capability is presented. The proposed submodule allows to operate with different internal capacitances in order to reduce the cost of the system. Furthermore, a control and balance strategy for the novel 5-level submodule is proposed.

In this work, a Model Predictive Control (MPC) strategy that combines Finite-Control-Set MPC (FCS-MPC) with Selective Harmonic Elimination (SHE) modulation pattern in its formulation is proposed to govern power converters. The key idea here is to define a desired steady-state in terms of the converter current and voltage. To do this, based on the converter current (system state) reference, an associated predefined SHE pattern is used as control input reference. Both system state and control references are included in the cost function. Therefore, during transients, the resulting predictive controller prefers to track the converter current while preserving the SHE pattern during steady-state. Thus, a fast dynamic response can be achieved throughout transients while a predefined voltage and current spectrum with low commutation frequency is obtained in steady-state by adjusting only the weighting factor of the cost function. Simulation results of the proposed SHE-MPC strategy when governing a Cascaded H-Bridge (CHB) converter are presented to highlight the benefits of the proposed predictive controller.

Direct torque control (DTC) scheme is quite popular in a voltage source inverter fed induction motor (IM) drive. Considering a switching table based algorithm, the flux and torque are directly controlled by applied inverter voltage vectors. There is a direct relation between stator flux and applied inverter voltage vector, which is eventually the stator voltage vector. However, in a current source inverter (CSI) fed IM drive, there is no direct relation between the stator flux amplitude and the applied current vector. Further, filter capacitor in between the inverter and motor terminals decouples the motor from the CSI. Hence, it is quite difficult to realise a switching table based algorithm to control stator flux. This paper proposes a new scheme to control flux and torque directly. The torque angle is calculated based on the electromagnetic torque demand; the torque angle and the output of flux controller generate the reference current vector. The dc link current is decided by the magnitude of the reference current and the CSI is switched through space-vector pulse width modulation scheme based on the the position of the reference current vector. The whole scheme is simulated in MATLAB/Simulink platform. The results show torque and flux are controlled properly.

This paper proposes a bipolar-dc-bus Electric Vehicle (EV) charging station without a line-frequency transformer at the medium-voltage grid side. It is mainly composed of the Neutral Point Clamped (NPC) converter based central ac-dc station and the parallel isolated three-level dc-dc converter based fast chargers. Due to the elimination of the line-frequency transformer, the volume and cost of the system are reduced. Meanwhile, the isolation and step-down between high dc-bus voltages and low load voltages are performed through the high-frequency transformers, leading to increased voltage utilization of switching devices. In order to guarantee the bipolar dc buses balance and minimize the dc-bus voltage ripples, the 180°-interleaved operation for the two parallel isolated three-level dc-dc converters is proposed to make the total neutral-point current zero so that all fast chargers do not bring any unbalance problems. The proposed 180°-interleaved operating principle has been analyzed in detail, then simulation results are presented to verify its performance.

This paper presents a comprehensive study on the state-of-the-art and emerging wind energy technologies from the electrical engineering perspective. In an attempt to decrease cost of energy, increase the wind energy conversion efficiency, reliability, power density, and comply with the stringent grid codes, the electric generators and power electronic converters have emerged in a rigorous manner. From the market based survey, the most successful generator-converter configurations are addressed along with few promising topologies available in the literature. The back-to-back connected converters, passive generator-side converters, converters for multiphase generators, and converters without intermediate dc-link are investigated for high-power wind energy conversion systems (WECS), and presented in low and medium voltage category. The onshore and offshore wind farm configurations are analyzed with respect to the series/parallel connection of wind turbine ac/dc output terminals, and high voltage ac/dc transmission. The fault-ride through compliance methods used in the induction and synchronous generator based WECS are also discussed. The past, present and future trends in megawatt WECS are reviewed in terms of mechanical and electrical technologies, integration to power systems, and control theory. The important survey results, and technical merits and demerits of various WECS electrical systems are summarized by tables. The list of current and future wind turbines are also provided along with technical details.

Common mode voltage reduction is one of the most important issues for transformerless PV grid-connected inverters. Many interesting single-phase inverter topologies have been reported to reduce the common mode voltage. However, three-phase PV inverter topologies with reduced common mode voltage are not well explored. In this paper, a novel three-phase seven-switch (H7) inverter is proposed to reduce the common mode voltage. It is derived from the conventional three-phase inverter, but has different operation principles. Therefore, a new modulation is presented to achieve the effective common mode voltage reduction. In addition, a simple control strategy is proposed to achieve the fast and accurate grid current regulation. Theoretical analysis and performance evaluation tests are carried out to verify the effectiveness of the proposed H7 topology and control strategy.

This paper presents a new seven-level voltage source converter (VSC) for high-power medium-voltage (MV) motor drives. The proposed topology is composed of H-bridge connection of two nested neutral-point-clamped (NNPC) converters called 7L-HNNPC converter. This converter has interesting features such as operating over a wide-range of output voltages particularly for a range of (10kV-15kV) without the need for connecting power semiconductor in series, high quality output voltage, less number of components in compare to other classical seven-level topologies. A novel SPWM technique has been developed for the proposed 7L-HNNPC converter to control and balance of the flying capacitors. The outstanding property of the control strategy is that the control can be applied independently to each leg of the converter that reduces the complexity of the controller significantly. The performance of the proposed converter is studied under different operating conditions in the MATLAB/Simulink environment. The feasibility of the proposed converter is evaluated experimentally on a scale-down prototype.

This investigation presents a grid-connected five-level H-bridge neutral-point-clamped converter operating with a single dc-link as photovoltaic power source. The operation with a single dc-bus is enabled by adding a transformer with open-end winding secondary, where each power cell of the converter is connected. The proposed control scheme is validated trough simulation results and tested under dynamic conditions. It can be implemented with a good tracking of the active and reactive power references injected to the grid. The results serve as a preliminary validation to potential utilization of the converter in large scale photovoltaic energy conversion systems.

A new space vector modulation method for common-mode voltage reduction in four-level nested neutral point clamped (NNPC) inverter is proposed in this paper. The proposed method employs switching states with minimal absolute common-mode voltages to synthesize given reference vectors. A variety of virtual vectors are constructed to facilitate the synthesis. The effect of dead time on common-mode voltage reduction is analyzed. Appropriate switching sequences are designed to avoid the effect. With the proposed method, the common-mode voltage of the inverter is reduced to 1/18 of dc-link voltage, 80% off compared with sinusoidal pulse width modulation (SPWM) and conventional space vector modulation (C-SVM), which contributes to the design of transformer-less medium voltage (MV) power conversion and drives. The proposed method is verified by simulation results.

In high-power medium-voltage applications, the device switching frequency of current-source converters (CSCs) is usually limited to several hundred Hertz. The selective harmonic elimination (SHE) is preferred due to its superior harmonic performance. However, the valid solution for the switching angles of SHE may not exist in some cases. The converter may switch to trapezoidal pulse-width modulation (TPWM) or space vector modulation (SVM) for solution. The TPWM and SVM schemes can provide variable modulation index control but the output current contains low-order harmonics with high magnitudes. In addition, compared with SHE and TPWM, the peak value of the common-mode voltage (CMV) could be doubled with conventional SVM because of the introduction of zero vector. In this paper, a novel SVM based modulation strategy is proposed for CSCs. Compared with the conventional SVM, the proposed method provides a lower CMV magnitude. In addition, the low-order harmonic contents can be reduced compared with the TPWM or the conventional SVM. Simulation and experimental results are provided to verify the proposed concepts.

This paper proposes a Machine Drive based on a Modular Multilevel Converter. The proposed converter uses current source cells with asymmetrical IGCTs as switching devices. A control scheme to manage the input and output current and, additionally, reduce the oscillations in the inductors currents is also proposed. The resulting machine drive will exhibit high power and high dynamical performance as shown in the simulation results.

This paper proposes a new power converter topology for the Wave Dragon multi-megawatt wave energy conversion platform. The current wave dragon power converter system is composed of several low-voltage two-level voltage source converters in back-to-back configuration; one per turbine-generator. Although this is a well known and proven technology, the use of distributed power electronic converters among all turbines does not bring any additional benefit in terms of efficiency and power quality. The proposed topology remains operating at low voltage and distributed at the generator side, but is combined at the grid side into a three-phase cascaded H-bridge (CHB) multilevel converter operating at medium voltage. This is achieved interfacing each PMSG through a low-voltage full bridge diode rectifier, followed by a boost dc-dc converter connected to the dc-link of the power cells of a 7-level CHB. The proposed configuration allows to concentrate the whole Wave Dragon platform into a single grid-tied converter with higher power quality and efficiency. Control schemes for both grid- and generator-side converters, and simulation results are presented in this paper.

The automotive industry is making important changes towards the electrified transportation. Several Electric Vehicles (EVs) and Plug-in Hybrid EVs are currently on the market offering a clean and environment friendly alternative to conventional vehicles. However, limited mileage capacity and long refueling times have lead to a lukewarm reception from the consumers. In order to reduce the charging time fast charging architectures are being developed, usually as off-board solutions located in public installations. This paper presents a centralized charging station architecture, using a bipolar dc grid to power off-board chargers and integrate of DGs and storage systems. The central converter topology is an NPC and a novel balancing technique is implemented to extend the operating range of the power converter.

A new predictive control strategy for current source rectifiers which allows an effective control of source and load currents is presented in this paper. This method uses the commutation states of the converter in the subsequent sampling time according to an optimization algorithm given by a cost function and the discrete system model. The two control goals are: (a) regulation of dc-link current according to an arbitrary reference, and (b) a good tracking of the source current to its sinusoidal reference. The feasibility of the proposed method is verified by MATLAB/Simulink software.

In this paper a new single-switch quasi- active power factor correction (PFC) converter based on two flyback dc/dc modules is proposed. To achieve high efficiency, part of the input power is directly transferred to the load by dc/dc flyback module 1. The rest of the input power is stored in the dc bus capacitor and reprocessed by a smaller dc/dc flyback module 2. The dc bus capacitor and the dc/dc module 2 also serve as a regenerative snubber. Due to the direct power transfer, lower voltage stress on the dc bus capacitor is achieved. A quasi-active PFC circuit is added to improve the power factor. The quasi-active PFC circuit is driven by a magnetic switch using a third winding coupled to the transformer of the dc/dc module 1. The input current harmonics of the proposed converter can easily meet IEC 61000-3-2 Class D requirements. The dc bus capacitor voltage is always equal the peak input voltage regardless of the load condition. Operating principles, analysis, and simulation results are presented to verify the effectiveness of the proposed converter.

In high power applications, back-to-back (B2B) current-source converters (CSCs) in direct parallel connection results in topologies that allow for transformerless configuration and multilevel current waveforms. However, the unbalanced dc-link currents in steady state and the current ripples in transient are inevitable due to the tolerance of devices and the pulse width modulation (PWM) switching pulse, which involve different voltage-drops across the dc-link inductors. The unequal voltage-drops introduce the unbalanced currents and ripples. This paper presents a way to reduce the dc-link current ripples based on the steady-state current balancing control, which is implemented by proper selection of the redundant switching states and the sequence design. The essence of the strategy is to reduce the voltage-drops across the dc-link inductors, as well as the time invertal of the switching states that involve the same dc link. The control scheme for direct parallel CSCs is implemented based on multilevel space vector modulation (SVM) algorithm. The proposed concepts are verified by a 2MW/4160V Matlab/Simulink model.

Modular multilevel cascaded converter emerges as an important alternative in high voltage DC transmission systems, due to its high reliability and power quality. This converter is composed by several identical cells whose are, in turn, composed by a chopper or an H-bridge and a floating DC capacitor. This paper proposes a modular multilevel converter with a novel cell topology based on a current source H-bridge. The use of this cell could improve the input and output power quality and, due to the IGCT current and voltage ratings, reduce the number of cells in series to achieve the required DC voltage and at the same time increase the transmitted power. Analysis and simulation of the proposed topology and its control are given.

This paper presents a novel model predictive control method for current source converters (CSC) following the duality with voltage source converters (VSC). In VSC the predictive control is applied to the output current, while in this work the CSC is controlled through the output voltage. The method is based on the discrete prediction model of the system, including the converter and output filter. The model is used to predict future behavior of the system variables for each switching state of the converter. The predictions are evaluated in a cost function that weights the level of accomplishment of the control goals. The switching state with best performance is then generated. The proposed algorithm predicts the output terminal voltages (includes the output filter capacitors), ensuring sinusoidal voltage generation without the need of cascaded control loops, coordinate transformations and modulation schemes found in classic solutions. In addition, low switching frequency is imposed by including the switching frequency in the cost function. Simulation results show a preliminary validation of the proposed control scheme, with comparable performance to the classical solutions, especially in terms of the current and voltage harmonic distortion.

Presented in this paper is a new strategy for indirect matrix converters that effectively mitigates common-mode voltages and allows an optimal control of source and load currents. This method uses the commutation state of the converter in the subsequent sampling time according to an optimization algorithm given by a simple cost function and the discrete system model. The control goals are regulation of output current according to an arbitrary reference and also a good tracking of the source current to its reference, which is imposed in order to obtain a sinusoidal waveform with low distortion. The technique is enhanced by a reduction of the common-mode voltage with an extra term in the cost function so as to avoid early motor winding failure and bearing deterioration. Simulation results support the theoretical development.