Search

This paper presents a method for designing the control loop for dc-to-dc converters when uncertainties exist in the ac characteristics of the converter's load. In the proposed method, a converter is initially considered as a stand-alone module feeding a current sink load and the control loop of the converter is then designed in a way that maximizes the robustness of the converter's closed-loop performance against the unknown ac dynamics of a potential load. As a result, the proposed control design method can provide the predictable closed-loop performance for a converter when it is loaded with an actual load whose ac characteristics are unknown in advance. Index Terms--Control design of dc-to-dc converters, dynamic performance of dc-to-dc converters.

A multi-input dc/dc converter based on the flux additivity is proposed in this paper. Instead of combining input dc sources in the electric form, the proposed converter combines input dc sources in magnetic form by adding up the produced magnetic flux together in the magnetic core of the coupled transformer. With the phase-shifted pulsewidth-modulation (PWM) control, the proposed converter can draw power from two different dc sources and deliver it to the load individually and simultaneously. The operation principle of the proposed converter has been analyzed in detail. The output voltage regulation and power flow control can be achieved by the phase-shifted PWM control. A prototype converter with two different dc voltage sources has been successfully implemented. Computer simulations and hardware experimental results are presented to verify the performance of the proposed multi-input dc/dc converter. Index Terms--Flux additivity, multi-input dc/dc converter, phase-shifted pulsewidth-modulation control.

We point out a thermodynamic interpretation of the variational Maxwell theorem relative to dc circuits, based on the minimum of the entropy production. We thus complete the interesting pedagogical analysis of D. A. Van Baak in a recent paper concerning the use of variational methods on dc circuits.

Neutral base recombination is shown to be the limiting factor controlling the maximum current gain of AlGaAs/GaAs heterojunction bipolar transistors (HBTs) with base doping in the 4 x 10 (super 19) negative cubic centimeters range at high injection levels. Results from the study also suggest that the minority carrier lifetime in all GaAs HBTs with base doping around 4 x 10 (super 19) negative cubic centimeters reported so far has been limited by recombination at trap sites in the quasi-neutral base.

In a generating station linked to a main system through an HVDC transmission system, a load rejection of the HVDC line may cause generators in the station to become self-excited, resulting in a severe system overvoltage. It may also cause damage to generator exciters if negative current is not allowed to flow in the exciter. This paper derives the self-excitation operating constraint for generators operating in such a system. Results show that a self-excitation operating constraint greatly reduces the operating zone of a generator in such a station. However, this operating constraint can be effectively improved by self-excitation protection systems to remove surplus AC filter capacitive VARs.

The analysis of random pulsewidth modulation (PWM) techniques has matured into a state where analytical expressions have been derived to aid in understanding the frequency domain characteristics. Derivations of the formulas require a lot of algebra, and the expressions must be verified by laboratory measurements. It is shown, however, that factors originating from digital signal processing techniques make the comparison difficult if proper measures against misinterpretation are not taken. A methodology to overcome the problems is presented. Novel expressions for the random lead-lag pulse position technique and the random switching frequency technique are also presented as well as their verifications by laboratory measurements on a full-bridge dc/dc converter. Index Terms - DC/DC converters, random pulsewidth modulation, spectral analysis.

This paper introduces fuzzy logic control as a technique for significantly improving the large-signal performance of paralleled de-de converters with master-slave current sharing control. The control design for such systems is challenged by its high modeling complexity, and present approaches come only from a small-signal perspective unable to give a good response for wide load and line transients. The fuzzy logic approach, by dealing naturally with nonlinearities, offers a good and parameter-insensitive transient response for the current sharing loop. Design methodology of the fuzzy compensator, simulation results, and a comparison with existing classical designs are shown.

One of the most important aims in the electronic industry is the integration of magnetic elements. since it allows the reduction of the size of power electronics equipment. This paper presents a method of designing integrated magnetic elements using thick-film technology. A dc-to-dc converter has been built and tested to validate this technology. Experimental results obtained and conclusions are also presented. Index Terms - DC-DC power conversion, magnetic components design, magnetic integration.

Problems involving the flow of direct current in circuits are typically solved through the use of Kirchhoff's rules - the junction theorem and the loop theorem. Since students typically find the meaning and use of the junction theorem straightforward, but are prone to conceptual and computational difficulties with the loop theorem, we present here a little-known alternative method for solving dc circuit problems. In this method, the junction theorem is retained, but the use of the loop theorem is replaced by the minimization of a single global variational quantity for the circuit. In this paper, the variational method is described and its use is illustrated, and its equivalence to standard methods is proved by vector calculus from the steady-state limit of Maxwell's equations.

New current-mode sinusoidal oscillator circuits are presented. The proposed circuits use single four-terminal floating nullor and, at most, eight passive elements. Two of the oscillator circuits use grounded capacitors and enjoy the independent grounded-element control of the frequency of oscillation and the condition of oscillation. Experimental results are included. Index Terms - Current-mode circuits, sinusoidal oscillators.

A study was conducted to analyze methods wherein an enhanced direct current short-circuit computation result can be determined by altering the fault current contributions from static sources. Ways for providing multiplying factors to adjust fault contributions related to static sources that differentiate between local and remote sources are examined. Results indicated that the simplest technique of enhancing the underestimate of fault currents is to partition the equivalent source resistance for a fault at the rectifier terminals by a factor of 1.15.

A simple circuit for direct current (dc) measurement using a transformer is proposed. The circuit is based on nonlinear effects in the transformer ferromagnetic core, and it provides information about the current magnitude and polarity. The circuit is analyzed applying piecewise-linear model of the core magnetization curve. An averaged-circuit model and the circuit transfer function are presented. Good linearity and wide measurement range are obtained. Design guides for reduction of the output voltage ripple are given. Measurement error within 0.25% of the full scale is achieved.

A physically based analytic DC model for analyzing the characteristics of asymmetric trapezoidal gate (ATG) metal oxide semiconductor field effect transistor (MOSFET) for both forward and reverse modes of operations is developed. In comparing the threshold voltage of ATG MOSFET with measurement data, it is noted that the model is in good agreement in both forward and reverse operation modes. Thus, the model can be used for circuit simulation using ATG MOSFET and can also be used for physical analysis of the ATG structure.

Soft-switching techniques have recently been applied in the design of dc-ac converters, in order to achieve better performance, higher efficiency, and higher power density. A substantial number of new topologies for different applications has been developed, however, the amount of work that has been done in this field is not widely known. This paper is an attempt to classify the soft-switched dc-ac topologies in a simple and generic way. The topology classifications are based on the location of the resonant network (load, inverter bridge, and bus), the characteristic of switching waveforms (zero-voltage switching or zero-current switching), and the type of resonance (series or parallel). Operating principles, performance, and design limitations are discussed. Some possible industrial applications of soft-switched dc-ac converters are addressed. Index Terms - Resonant dc-ac converters, soft-switched dc-ac converters, zero-voltage switching, zero-current switching.

This paper considers the sequence impedance and equivalent circuit of an HVDC converter and system. The results can be used in the steady state analysis of power systems containing HVDC terminals with systems operating under balanced or unbalanced conditions. The effects of commutation overlap are considered. The method described provides a simple model for steady state studies for power systems containing converters and HVDC links.

The design of high-power inductors for applications with broad current spectrum excitation is a challenging task. The resonant inductor of a resonant dc-link inverter (RDCLI) is one such example. The inductor current consists of a resonant current component, a dc component, which supplies the active power to the load and a modulation component, which depends on the modulation strategy. In addition, the frequency and amplitude of the dominant current components change with operating point. Conventional inductor designs for single-frequency excitation do not perform well in broad-spectrum applications. In order to improve these designs, the impact of broad current spectrums on winding design, core selection, power density, and thermal-handling capability must be investigated. In this paper, alternate inductor topologies, which better address the above issues, are proposed and investigated. Index Terms - High-frequency magnetics, inductor design, resonant dc-link inverter, strip windings.

The design of an energy-storage reactor in large measure determines the performance of most high-efficiency switched-mode dc-to-dc converters. No other component has such a dramatic effect on the circuit waveforms and the distribution of circuit losses. It is often difficult, even for experienced designers, to visualize the complex relationships between the choices of magnetic design parameters and the resulting electrical performance of the power stage. This paper develops a methodology that allows the designer to relate the magnetic design choices of core material, core geometry, number of turns, and relative permeability to the electrical performance of the power stage. Index Terms - Converter, dc-to-dc converter, design, energy-storage reactor, inductor, magnetics, switch-mode supply.

In this paper, the authors verify the calculation procedures proposed in the International Electrotechnical Commission (IEC) draft Standard for the evaluation of short-circuit currents in dc auxiliary systems of power plants and substations. The dynamic simulation performed on test networks by the Alternative Transient Program (ATP) is taken as a reference for a comparative analysis of the two approaches. Index Terms - DC systems, standards, short-circuit currents.

This paper is concerned with the macromodeling of a 'switch' which is important from the point of view of SPICE simulation of dc-dc converters, switching mode power supplies, and power electronic circuits. In this paper, some new macromodels of switches are introduced which are simpler and allow faster simulation times than a previously known model. The underlying principle and the mechanism of the proposed models has been explained in a manner suitable for educational purposes. Index Terms - Macromodeling, switch modeling, switch simulation.

Low-grade exhaust heat is used to provide a reliable and independent power source for instrumentation circuitry by means of a thermoelectric generator (TEG). A design of a self-starting dc-dc converter is developed and optimized for very-low-input voltages (below 300 mV) in order to allow operation at temperature differences of 20 [degrees] C and less. A prototype is built, and the results are experimentally verified. Index Terms - DC-DC power conversion, losses, low power, low voltage, power generation, self-starting, thermoelectric generator, voltage control.

This paper presents a new topology for a pulsewidth modulation (PWM) rectifier which achieves unity power factor on the ac supply side and ripple reduction on the dc output side. The main circuit of this rectifier consists of a conventional PWM rectifier and a pair of additional switches. The switches and PWM rectifier are controlled such that the ripple current on the dc line is reduced, and unity power factor is achieved on the ac line. As a result, this circuit does not require a large dc capacitor or a passive LC resonant circuit. Furthermore, control of the additional switches and PWM rectifier requires only a simple control circuit. The effectiveness of this circuit was confirmed by experiments and analysis. The rectifier is useful for uninterruptible power systems (UPS's) and dc power supplies, especially for cases in which batteries are connected to the dc line. Index Terms - Battery, PWM converter, ripple.

High-current dc thyristor circuit breaker (t.c.b.) devices, based on the parallel capacitor-current commutation-forced turn-off principle, require very large capacitors to effect correct operation as fault-current interrupters in electrical circuits. The commutation capacitors of t.c.b. devices that interrupt current in a single switching operation absorb virtually all the electromagnetic energy stored in circuits, where they operate during the single switching process. These device types interrupt current extremely fast, but because of the scale of the energy transfer to the commutation capacitor, they are especially prone to producing short-duration switching overvoltages of excessive magnitude. A novel t.c.b. device involving two switching operations or stages is presented and analyzed. With this device type, the circuit-stored energy is transferred to the commutation capacitor in two stages, which, in so doing, produces significantly lower switching overvoltages than the conventional 'single-stage' t.c.b. device. Results of a range of tests on a low-voltage two-stage t.c.b. device and a high-power-prototype version are given together with the basis for determining their optimum design criteria. Comparative simulation studies are also provided. Finally, the limitations of t.c.b. devices are examined and discussed in conjunction with cost comparisons and their merits and limitations for typical applications. Index Terms - DC circuit breakers, dc electrical-protection systems, dc-traction systems, thyristor circuit breaker, thyristor switch.

Concepts of maximum power transfer of dc systems and associated ac voltage variations, particularly at inverter stations having low short - circuit ratios, have been extended to include various ac load models and an ac line in parallel with the dc line. The operating capabilities are shown to vary from those predicted from either a Thevenin ac source model or the corresponding short-circuit ratio. The study used an ac/dc load flow program. Keywords - HVDC, short-circuit ratio, power transfer, ac/dc interaction.

A new procedure is presented that very rapidly determines whether or not a dc operating point exists for a series-parallel network of continuous, strictly monotonically increasing resistances having possibly limited domains and/or ranges. This is accomplished without any attempted computation of the operating point. If such exists, it will be unique. The procedure does not involve any manipulation of the characteristic curves of the resistors other than the asymptotes to which they tend. As a result, it is easily programmed and rapidly executed. The test is extendable to series-parallel networks of nonmonotonic resistors, but in this case the test is only for existence - not for uniqueness. Index Terms - Monotone resistors, nonlinear networks, operating points, series-parallel networks.

An accurate model of a thyristor in dc chopper commutation circuits is described using the available elements in the PSpice program library. The improved model, which takes into account the reverse recovery time during turn off, is used to simulate voltage and current commutated dc choppers. Simulation results, which include a study of commutation failure, agree with theoretical and experimental results. The model developed is useful for computer aided analysis and design of dc thyristor choppers including the commutation circuits.

The Class-E dc-dc converters with half-wave and full-wave controlled current rectifier are proposed. Zero voltage switching for all the switches can be maintained from full load to no load. The switching frequency is constant. When the load current changes from maximum to zero, the output voltage can be kept constant by regulating the conducting angle of the rectifier switch. The operation of the new converters is analyzed and the zero voltage switching characteristic has been demonstrated. The current gain of the circuit versus the conducting angle has been presented.

Low-voltage, direct current equipment in utility control systems, uninterruptible power systems (UPS), telecommunication systems, and elsewhere has received increasing attention in recent years. Power conditioning needs are one reason. Closer attention by regulatory agencies is another. This paper is intended to help fill a gap in available material on ratings and application characteristics of low-voltage circuit breakers in dc systems. Some of this material is prepared for use by the IEEE Battery Protection Working Group in developing a Guide for Protection of Stationary Battery Systems under project P1375.

A fast and convergent iteration method for piecewise-linear analysis of nonlinear resistive circuits is presented. Most of the existing algorithms are applicable only to a limited class of circuits. In general, they are either not convergent or too slow for large circuits. The new algorithm presented in the paper is much more efficient than the existing ones and can be applied to any piecewise-linear circuit. It is based on the piecewise-linear version of the Newton-Raphson algorithm. As opposed to the Newton-Raphson method, the new algorithm is globally convergent from an arbitrary starting point. It is simple to understand and it can be easily programmed. Some numerical examples are given in order to demonstrate the effectiveness of the proposed algorithm in terms of the amount of computation.

In force-commutated SVC and DC Transmission stations, series-connected, multi-converter modules are necessary to build up the dc link voltage rating. Based on digital simulations of 8 modules, it is shown that the Total Harmonic Distortion (THD) of 5% can be achieved from harmonic cancellations by using phase shifting of the triangular carriers in the Sinusoidal Pulse Width Modulation technique. The requirement of the dc link capacitor size to ensure ripple-free dc voltage is studied. The 8-module case-study points to the ability to achieve 1% THD for higher voltage rating by increasing the number of modules connected in series.

This paper presents new results concerning the stability of operating points of dc circuits. We define two classes of unstable operating points and develop criteria for identifying the class to which a given operating point belongs. We also show, by employing results from degree theory, that if a circuit possesses n structurally stable operating points (n has been shown previously to be odd), then 1/2(n - 1) of these operating points must be unstable and hence physically unobservable. A special case of this result proves that any bistable circuit must posses at least three operating points.

With the advent of faster microprocessors, refinements in the control systems for thyristor-fed dc drives have become possible. In such drives, the problem of detecting the true zero crossings of the input voltage, when the mains supply is contaminated with noise spikes, is quite important. A new approach for this problem, based on the predictive synchronization concept, is presented in this paper. The algorithm presented also covers the other aspects, like generation of firing pulses and closed loop calculations for the drive control. The microcomputer based system has been tested on a separately excited dc motor and the test results are also presented in this paper.

When an HVdc scheme is isolated from its receiving ac system, the inventor may continue to operate, generating its own ac bus voltages; this is defined here as islanding. If islanding is allowed to continue unrestricted, then main circuit components may in some conditions be damaged and it is therefore necessary to provide a suitable protection system. This paper outlines the protection scheme developed for the McNeill Back-to-Back HVdc link in Alberta, Canada, to prevent damage due to islanding while still permitting the link to automatically restart on reclosure of the isolating breaker. Oscillograms showing the protection in operation on both the GEC ALSTHOM HVdc simulator and during tests carried out as part of the commissioning of the McNeill HVdc link are included.

The purpose of introducing active dc filters is to meet the more and more stringent requirement from power utilities on limiting telephone interference caused by harmonic currents from HVdc transmission lines, without unnecessarily increasing the cost of HVdc stations. An active dc filter installed in the Konti-Skan HVdc link is described. The active dc filter is connected at the bottom of an existing passive dc filter at the Lindome station. The active dc filter includes optic harmonic current measuring unit, control system, protection and supervision system, PWM power amplifier, high-frequency transformer, surge arrester, and coupling apparatuses. The active dc filter has small physical size and occupies small ground area. The performance of the active dc filter for eliminating the disturbing harmonics is excellent. To achieve comparable results by passive filters would require something like ten times more high voltage equipment.

Bipolar dc corona is known to be a strong function of conductor voltage but little has been known about the influence of load current on bipolar corona. To evaluate the influence of load current, a bipolar cylindrical corona cage was constructed and operated with simulated dc load current flowing in the conductors. Measurements of load current, corona current, ion-current density, and conductor temperature showed that corona conductance was a function of load current. The dependence on load current was via temperature rather than magnetic field.

The experimental and analytical current-voltage characteristics of four types of direct current superconducting quantum interference devices with different parameters of coupling section were studied. The mechanism and damping condition for resonant peak in the devices' were then evaluated based on the devices' current-voltage properties. A strong agreement is observed between the experimental and analytical results.

Considerations for direct current design in relation to the optimal epitaxial silicon- and silicon-germanium (SiGe)-base bipolar transistors for liquid nitrogen temperature (LNT) conditions are analyzed. Results show that the thin and abrupt base profiles processed epitaxially are conducive to low-temperature operations. LNT conditions also encourage a vertical profile with lightly doped spacer layers and a graded-bandgap SiGe base.

A new method of modeling PWM converters operating in continuous mode (CCM) is introduced. The procedure of deriving the models consists of three steps. First, static voltage and current transfer functions of the idealized switching part of the converters are found. Second, the linearization of these transfer functions at the operating point is carried out and the idealized switching part is replaced by dependent current and voltage sources. Third, the equivalent average resistance (EAR) of parasitic resistances and equivalent average voltage (EAV) of offset voltage sources of switches are determined using the principle of energy conservation. The method leads to linear dc and small-signal circuit models of a PWM converter. The obtained models give the same dc and small-signal characteristics as the state-space averaging method. To illustrate the method, the analysis of the PWM flyback converter is given, taking into account parasitic resistances of the transformer, the filter capacitor, the switch, and the diode as well as the diode threshold voltage. Design equations for dc voltage transfer function, efficiency, and small-signal characteristics are derived.

A new design for a first generation current conveyor implemented from complementary bipolar transistors is introduced. Its input cell consists of two translinear mixed loops in parallel. The circuit that acts in AB class can process currents with magnitudes greater than the bias current [I.sub.o] and so generate very low harmonic distortion. It is characterized by low values for its parasitic input resistance and large -3 dB bandwidths. It can also be used with supply voltages as low as [+ or -] 1.2 V. Index Terms - Bipolar transistors, current conveyors, current mode circuits, low voltage, low power, translinear circuits.

The usefulness over an extended range of a HEMT model previously validated for 1 to 25 GHz s-parameter model is shown. Experimental and simulation results for the dc drain current and 1-50 GHz s-parameters of a pseudomorphic 0.32-micrometer-gate AlGaAs/InGaAs/GaAs HEMT are presented. The model predicts the device's dc current and s-parameters as a function of the applied gate bias with good accuracy. The core of the model is directly dependent on the HEMT wafer structure and the physical gate length. As part of the modeling procedure, a value of (1.77 + or - 0.07) X 10 to the fifth power m center dot s to the -1 is found, confirming the results of other research, for the electron velocity in undoped pseudomorphic In0.15Ga0.85As under approximately or equal to 0.3 micrometer gates.

A pulsewidth-modulated (PWM) dc-dc converter enters discontinuous conduction mode (DCM) as its inductor current attempts to reverse direction, but cannot do so because of switch implementation, and remains at zero in the last interval of each switching period. Three equivalent three-terminal nonlinear transformers are described as the models for the PWM switch in a DCM converter, each having one of the three terminals serving as the common terminal. Each model facilitates the analysis of those corresponding converters in which the PWM switch terminal corresponds to the common transformer terminal is dc-grounded. Examples are given to demonstrate how the models are used to derive the DCM criteria, and the ideal and nonideal voltage gains for the buck-boost, buck, and boost dc-dc converters. The results predicted by the models agree with those reported and verified in the literature.

This paper develops a constructive homotopy-based methodology for finding all or multiple dc operating points of nonlinear circuits and systems. Several sufficient conditions for the connectivity of all the solutions along a single homotopy path are derived. These conditions offer criteria to determine an initial guess from which all the solutions can be obtained by following the single homotopy path. For the class of nonlinear circuits and systems in which all of the solutions lie on several homotopy paths, a new systematic method to explicitly construct a starting point for each homotopy path is developed. From a practical viewpoint, the constructive method developed does not require the difficult task of finding a good initial guess and is applicable to general nonlinear circuits and systems. From a methodological viewpoint, the constructive method developed is applicable to general homotopy methods with different homotopy functions. One particularly important feature of the developed constructive method is that when this method is applied to a dc problem with box constraints, it is very computationally efficient because it locates only feasible solutions satisfying the given box constraints. The proposed method is applied to several test examples with promising results. Index Terms--DC operating points, homotopy methods, nonlinear circuits and systems.

This paper discusses the causes of and some solutions to the commonly observed problem of dc field offsets in finite-difference time-domain (FDTD) simulations. DC electric and magnetic field offsets are shown to be valid calculated responses of the modeled systems, resulting from interaction between the turn-on characteristics of the source and the properties of the models. The dc offsets may be avoided in the time domain by tailoring the source waveforms or in the frequency domain by post-processing the FDTD output. Index Terms--Finite-difference time-domain (FDTD) methods.

The switch-mode dc-to-ac inverter using dc-to-dc converter configuration has many advantages over the traditional bridge-type inverter. The previously presented circuit with nonlinear robust control works only for resistive load and its output voltage has considerable distortion near the zero-crossing point. A novel bidirectional circuit with nonlinear robust control is proposed in this paper to overcome those drawbacks. This inverter can dynamically stabilize the output voltage, i.e., the output voltage of the inverter remains dynamically unchanged when subjected to large disturbances in the input voltage or output load. The stability analysis is presented. The closed-loop control system under P control, PI control, PD control, and PID control of the voltage loop for different types of loads are investigated. The results show that the PD control is the best choice. The performance of the circuit is also studied. Computer simulation and experimental results are provided. Index Terms--Control techniques, dc-ac power conversion, dynamic response, pulse width modulation inverters.

A novel inductor voltage control (IVC) method capable of achieving near unity power factor is being proposed for buck-type ac-dc pulsewidth-modulated (PWM) converters. In this method, the input inductor voltages are kept within a hysteresis band about a sinusoidal template, thus ensuring sinusoidal input currents. This control method is much less sensitive to parameter and control variations than the existing delta modulation control (DMC) method. A companion paper has introduced the IVC method for the case of a single-phase buck-type converter. In the present paper, problems involved in direct extension of the IVC method to three-phase buck converter are discussed first. Following this, a new switching logic scheme is proposed which enables these problems to be overcome and for IVC (as well as DMC) to be extended to a three-phase converter as well. Detailed simulation and experimental results have been provided to verify the expected good performance with IVC. The proposed IVC method has potential in applications such as those requiring ac--dc rectifiers with current limiting. Index Terms--AC-DC power converter, PWM rectifier, unity power factor rectifier.

Protecting transmission lines frequently involves adopting distance relays. Protective relays must filter their inputs to reject unwanted quantities and retain signal quantities of interest. Accuracy and convergent speed of filter algorithm are essential for protective relays. A widely applied filter algorithm, the discrete Fourier transform (DFT) can easily erase harmonics using simple calculation. However, the voltage and current signals contain large harmonics and dc offset during the fault interval. The dc offset heavily influences the precision and convergence speed of fundamental frequency signal from DFT. In this investigation, we present a novel Fourier algorithm to remove the dc offset in a voltage or current signal. Applying full-cycle DFT (FCDFT) requires one cycle plus two samples to calculate and compensate for the dc offset. Half-cycle DFT (HCDFT) only requires half of a cycle plus two or three samples to accomplish the algorithm when the input signal has no even order harmonics. Adopting the proposed algorithm in distance relays effectively suppresses the dc offset and quickly decomposes the accurate fundamental frequency components. Index Terms--digital filtering, digital relaying, distance relay, protective relays, suppression of dc offset effects.

Multilevel converters, like neutral-point-clamped inverters [5], [6] or multilevel choppers [3], are particularly attractive in high-power applications. Nevertheless, in these structures, all switches are confronted to commutation stresses caused by their turn-on and turn-off control. Furthermore, the methods to balance the capacitor voltages or to control the neutral point voltage are complex enough. In this paper, the authors propose new multilevel converters based on series connection of zero-current-source (ZCS) inverter cells and parallel connection of zero-voltage-source (ZVS) inverters. These dual structure associations give soft-switching operation for all switches [2], [4] and allow the use of semiconductors, normally destined for medium-power applications, in high-power converters (up to 1 MW). The authors consider the structure design for several topologies to achieve dc-dc or dc-ac converters. The simulation results validate the simplicity of phase control techniques and give out the principal features of different topologies. Index Terms--Duality, multilevel converters, phase control, ZCS mode, ZVS mode.