Semiconductor Losses Research Articles

Three‐mode pulse‐width modulation of a three‐phase four‐wire inverter

In order to reduce the switching frequencies of power semiconductors and then to increase the efficiency in a three-phase four-wire inverter, the state controllability of a system is investigated by considering the inverter as a switched linear system. It is found that a three-phase four-wire inverter is a three-mode controllable, therein, only three non-zero vectors are required to make the system completely state controllable. With the three operating modes, a novel control strategy, called a three-mode pulse-width modulation (3-MPWM), is proposed in this study. There are three types of 3-MPWMs, that is, 3-MPWM(I), 3-MPWM(II) and 3-MPWM(III). Compared with the conventional space-vector pulse width modulation strategy, the switching frequencies of the proposed 3-MPWMs are significantly reduced by two-thirds, hence, the switching losses of the power semiconductors can be reduced and the efficiency of the inverter can be improved. In addition, the number of operating modes used in 3-MPWM(I) and 3-MPWM(II) is reduced by a half. Finally, simulation and experimental results are conducted to verify the effectiveness of the 3-MPWM strategies.

Automatic Temperature Regulation System of Locomotive Traction Induction Motors With Power Losses Minimization

The air cooling systems are shown to be used to provide required temperature condition of traction induction motors on locomotives. The automatic temperature regulation system is developed for its using to solve such a task. Results of experimental investigation showed that the AO63-4 induction motor stator end winding on the side opposite to air supply is the most heated part of the induction motor. Based on the results of the research, it was used an aperiodic second-order transfer function for approximation of the thermal transient curves. The design of an induction motor control system maintaining operating mode with minimum of the stator current are considered. It is shown that the modes of minimum of the stator current and minimum of power losses are quite close to each other. The MatLab simulation results taking typical nonlinearities and iron power losses in an induction motor and conduction and commutation power losses in semiconductors of frequency converter into account are presented. It is shown that as a result of application of the suggested system the power losses reduction may be led up to 20 % relatively to classical scalar control.

Comparative Analysis of Semiconductor Power Losses of Galvanically Isolated Quasi-Z-Source and Full-Bridge Boost DC-DC Converters

Abstract This paper compares semiconductor losses of the galvanically isolated quasi-Z-source converter and full-bridge boost DC-DC converter with active clamping circuit. Operation principle of both converters is described. Short design guidelines are provided as well. Results of steady state analysis are used to calculate semiconductor power losses for both converters. Analytical expressions are derived for all types of semiconductor power losses present in these converters. The theoretical results were verified by means of numerical simulation performed in the PSIM simulation software. Its add-on module “Thermal module” was used to estimate semiconductor power losses using the datasheet parameters of the selected semiconductor devices. Results of calculations and simulation study were obtained for four operating points with different input voltage and constant input current to compare performance of the converters in renewable applications, like photovoltaic, where input voltage and power can vary significantly. Power loss breakdown is detailed and its dependence on the converter output power is analyzed. Recommendations are given for the use of the converter topologies in applications with low input voltage and relatively high input current.

Comparative loss analysis and efficiency performance of resonance-based PDP sustain drivers

Sustain drivers based on the LC resonant operation have a significant influence on the entire circuit efficiency and cost of plasma display panels (PDPs). Although circuit efficiency of PDP drivers is extremely important, theoretical loss analysis and comparative studies of the sustain drivers have been missed in the literature, despite a variety of developed PDP sustain drivers. This paper presents in detail theoretical analyses and systematic comparisons of semiconductor losses in the PDP sustain drivers, including the inevitable conduction losses by the displacement currents, the undesirable extra conduction losses, the switching losses by the forced current commutations and the switching losses by the step voltage changes. Based on derived theoretical analysis and comparative features, sustain drivers are evaluated in terms of PDP operating speed as well as circuits’ efficiency.

Hybrid Design of Modular Multilevel Converters for HVDC Systems Based on Various Submodule Circuits

The modular multilevel converter (MMC) has become the most promising converter technology for high-voltage direct current (HVDC) transmission systems. However, similar to any other voltage-sourced converter-based HVDC system, MMC-HVDC systems with the half-bridge submodules (SMs) lack the capability of handling dc-side short-circuit faults, which are of severe concern for overhead transmission lines. In this paper, two new SM circuit configurations as well as a hybrid design methodology to embed the dc-fault-handling capability in the MMC-HVDC systems are proposed. By combining the features of various SM configurations, the dc-fault current path through the freewheeling diodes is eliminated and the dc-fault current is enforced to zero. Several MMC configurations based on the proposed hybrid design method and various SM circuits, that is, the half-bridge, the full-bridge, the clamp-double, and the five-level cross-connected SMs, as well as the newly proposed unipolar-voltage full-bridge and three-level cross-connected SMs, are investigated and compared in terms of the dc-fault-handing capability, semiconductor power losses, and component requirements. The studies are carried out based on time-domain simulation in the PSCAD/EMTDC software environment for various SM configurations and dc-fault conditions. The reported study results demonstrate the proposed hybrid-designed MMC-HVDC system based on the combination of the half-bridge and the proposed SM circuits is the optimal design among all evaluated systems in terms of the dc-fault-handing capability, semiconductor power losses, and component requirements.

Hybrid Multilevel Converter With Cascaded H-bridge Cells for HVDC Applications: Operating Principle and Scalability

Hybrid multilevel converters are contemplated in an attempt to optimize the performance of voltage source converters in terms of magnitude of semiconductor losses and converter footprint, and to achieve additional features such as dc short circuit proof, which is essential for a high integrity multiterminal HVDC grid. Therefore, this paper considers an emerging hybrid cascaded converter that offers the dc side short circuit proof feature at reduced loss and footprint compared to the existing multilevel and other hybrid converters. Its operating principle, modulation, and capacitor voltage balancing strategies are described in detail. Furthermore, hybrid converter scalability to high voltage applications is investigated. The validity of the modulation and capacitor voltage strategy presented are confirmed using simulation and experimentation. The hybrid cascaded converter is extendable to a large number of cells, making it applicable to high voltage applications, and operation is independent of modulation index and power factor. On these ground, the converter is expected to be applicable for both real and reactive power applications.

Time constant measurement for control of induction heating processes for thixoforming

In controlling induction heating systems, several measurement methods exist for controlled heating of metal billets into the semi-solid state for thixoforming. The most common approach is to measure the billet temperature, which suffers from various drawbacks leading to difficulties in process stability. The main disadvantages are the small temperature range of the process window and the alloy composition dependency of the correlation between temperature and liquid fraction. An alternative is to determine the liquid fraction of the billet by measuring the time constant of the load. Although time constant measurement is not affected by the mentioned problems, it is difficult to use it as a controlled variable. This paper shows that disturbances affecting time constant measurement are mainly caused by semiconductor losses inside the inverter. A method is introduced to compensate these losses. This method was implemented and tested in the embedded system of an induction heating unit, thereby showing that it is possible to use time constant measurement to determine the liquid fraction of a billet during induction heating.

Comparison of power losses, current and voltage stresses of semiconductors in voltage source transformerless multilevel inverters

The power losses, current and voltage stresses in semiconductor devices in voltage source transformerless multilevel inverters (VSTMLIs) topologies are unequal, depending on the switching states, the duty ratio, the levels of output voltage, the load power factor and the depth of modulation index. Non-uniform power loss profiles mean both unequal junction temperature and power rating in semiconductors. In this study, a comparison of power losses, current and voltage stresses in semiconductors of VSTMLIs is performed. Initially, the structural characteristics and operating principles of dominant VSTMLIs with sinusoidal pulse width modulation are analysed. Then, modifications of existing VSTMLIs and simplified multilevel topologies are proposed, which achieve a further reduction in switch count and power losses in semiconductors. The proposed modified circuitries are suitable for converting existing four-level VSTMLI topologies into five-level ones. This conversion is desirable, when a zero output voltage level under no-load conditions or low modulation indices should be obtained. Power losses in semiconductors are calculated analytically and then confirmed by simulation. A comparative presentation of profiles of power losses, current and voltage stresses in semiconductors of the examined VSTMLIs is analytically carried out. The obtained results are valuable for sizing and selecting the semiconductors in VSTMLIs.

Diode-Free T-Type Three-Level Neutral-Point-Clamped Inverter for Low-Voltage Renewable Energy System

In this paper, the conventional I- and T-type three-level neutral-point-clamped (3L-NPC) inverter for low-voltage renewable energy systems is first investigated. Literature research shows that the T-type inverter improves I-type's insulated-gate bipolar transistor (IGBT) $+$ IGBT current paths. However, the IGBT $+$ diode paths are the same. The calculation in this paper further reveals that the IGBT $+$ diode current paths dominate the conduction losses and even the whole semiconductor losses. Based on the aforementioned recognitions, a novel T-type inverter is presented as an alternative to be applied in the low-voltage renewable energy systems. In the proposed 3L-NPC, four CoolMosfets replace the IGBT $+$ diode middle bidirectional switch. In the proposed topology, there is no diode involved in the current path supposing the unity power factor. In this way, the conduction loss is expected to be reduced compared with that from the conventional T-type 3L-NPC, particularly in the low and medium power ranges.

Single‐phase ac–dc–ac multilevel five‐leg converter

In this study, a multilevel five-leg single-phase ac–dc–ac converter and its generalisation are presented. The converter is suitable to be used as an uninterrupted power supply or a unified power quality conditioner without an isolation transformer. The operating principles, the pulse-width modulation strategy and the control system for the proposed converter are presented. The proposed single-phase multilevel ac–dc–ac converter is composed mostly of low-power switches capable of generating input and output voltages with low harmonic content. The proposed configuration is compared with the conventional solution in terms of harmonic distortion, semiconductor losses and ratings of switches. Experimental results demonstrate the feasibility of the proposed configuration. The experimental verification of the proposed topology and control strategy was obtained by using IGBTs with dedicated drives and a digital signal processor TMS320F28335 with plug-in boards and sensors.

A High-Efficiency Bidirectional AC/DC Topology for V2G Applications

This paper proposes a single-phase bidirectional AC/DC converter topology applied in V2G systems, which consists of an inverter and a bidirectional non-inverting buck-boost converter. This topology can operate in four modes: buck charging, boost charging, buck discharging and boost discharging with high input current quality and unity input power factor. The inverter switches at line frequency, which is different from conventional voltage source inverters. A bidirectional buck-boost converter is utilized to adapt to a wider charging voltage range. The modulation and control strategy is introduced in detail, and the switching patterns are optimized to reduce the current ripple. In addition, the semiconductor losses are analyzed. Simulation and experimental results demonstrate the validity and effectiveness of the proposed topology.

Power Quality Improvement in ac Railway Substations: The concept of chopper-controlled impedance

With a length of 9,698 km, the 25 kV/50 Hz AC single-phase supply is a widely used railway system in France. Overhead lines are supplied by substations drawing power from two phases of a 3-phase utility. They behave as nonlinear and time-varying loads and represent one of the most important sources of voltage unbalance for the electricity transmission network. In the case of weak networks, railway operators are required to install compensation systems in substations in order to satisfy utility regulations and to avoid penalties regarding voltage unbalance and reactive power consumption. The limits are established by the energy provider with a view to guaranteeing an acceptable power quality to other customers.In three-phase networks, widely used solutions are the classical SVC, based on Thyristor Controlled Reactors (TCRs), and STATCOM based on Voltage Source Inverters. The TCR allows the variation of fundamental lagging current by phase-control, counterbalancing large leading currents from associated fixed capacitors and allowing continuous compensation of the lagging line. However this solution generates a high level of harmonic and requires onerous LC filters. On the other hand, Voltage-sourced inverters (VSI), offer several advantages over thyristor-based solutions in terms of compensation dynamics and reduced harmonic distortion. For the last ten years, although it requires bulky capacitors on the DC bus, the VSI approach has been widely used for avoiding system unbalance. However, in high-power applications, the semiconductor losses bring forth significant costs both in terms of active energy and cooling system maintenance which must be taken into consideration by the railway operator. The concept of the Chopper-Controlled Impedance (CCI), presented in this paper, is based on series or parallel associations of AC Choppers using high-frequency pulse-width modulation to vary reactances at the network frequency. This approach is a low power-loss solution and requires a low volume of reactive elements, a fact which makes this solution very attractive in high-power single-phase systems such as railway networks.

SiC Power Devices in a Soft Switching Converter, Including Aspects on Packaging

In many applications of power electronic converters efficiency and size are important figures of merit. Low losses in the power semiconductors and high frequency operation are important factors. The converters considered in this paper are off-line powered and in the power range of 50 – 150 kW with a high-voltage dc-output. The switching frequency is 20-40 kHz. With modern IGBTs an efficient operation (n> 95%) can be obtained when using soft-switching topologies. Fig 1(a) shows the schematic diagram of an SLR, series loaded resonant converter [1]. In Fig 1(b) typical waveforms of transformer current and voltage are depicted. In a recent paper, [2], frequency dependent IGBT losses are evaluated. Fig 2 shows the on-state voltage of an IGBT when conducting a sinusoidal current at different frequencies. At the higher frequency (30 kHz, Tr4) the increase of conduction losses is evident. Power switches made from SIC show excellent properties enabling; low conduction losses, high frequency operation and an ability to operate at high die temperatures [3]. Fig 3 (a) shows the on-state voltage of an unipolar SIC switch while Fig 3(b) shows the on-state voltage of a bipolar switch. Measurements have been performed up to 200 kHz without showing any high-frequency effects. High voltage SiC PiN diodes can replace several series connected Si diodes due to the higher critical field strength in SiC. This enables substantially reduced conduction losses in the output rectifier. In Fig 4 on-state voltages of 10 kV SiC and Si-diodes are shown, indicating a 50% reduction of losses.

Design of Boost Converter Based on Optimum Weighted Average Efficiency for Photovoltaic Systems

This paper presents different optimum designs of DC/DC boost converters applied to photovoltaic systems (PV). Boost converter is used to track the maximum power of PV systems. The main goal of this work is to select of an operation point for this converter (current ripple @ switching frequency) and choosing the magnetic core and conductor diameter, which provide better efficiency in all load range. The losses in capacitors, semiconductors and magnetics are evaluated for all load range. After that, it is obtained the efficiency in 5%, 10%, 25%, 50%, 75% and 100 % of nominal power in order to calculate the weighted average efficiency, as determined by standard IEC 61683:2000. From this evaluation, the operation point and the magnetic material that will result in the best efficiency are selected. The volume and cost are not considered in this work. Experimental results are presented to validate the obtained simulated results, as well as results of the total efficiency of the boost converter. Four different configurations are presented to prove that the operating point and the magnetic core selected are the best among them. A discussion of the results is presented, where alternatives to improve efficiency throughout the power range are analyzed.

Quantitative Analysis of the Optical Losses in CZTS Thin-Film Semiconductors

For the first time, a quantitative approach was carried out to model the optical losses in several Cu2 ZnSnS <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> (CZTS) based thin-film heterojunction solar cells having different window and buffer layers. The modeling indicates that the heterojunction of glass/FTO/TiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> /In <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> /CZTS has a lower reflection and higher transmission coefficient than glass/FTO/In <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> /CZTS and conventional glass/FTO/CdS/CZTS structures. These optical losses at the interfaces (reflection) and inside the thickness (absorption) of stacked layers were calculated based on the known optical constants such as refractive index and extinction coefficient. While TiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> windowed device resulted in a promising high short-circuit current and energy conversion efficiency, the In <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> windowed and conventional CdS windowed devices showed to be less efficient. We compared our theoretical analysis with the practical reports given for such structures in the literature. Further studies are required to reduce the reflection and absorption losses of CZTS structures in practice. So far, the overall performance parameters (e.g., Jsc and η) are much less than our theoretical estimations. The size of the TiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> nanoparticles on the layer is about 100 nm.

A Single-Phase Current-Source Bidirectional Converter for V2G Applications

In this paper, a single-phase current-source bidirectional converter topology for V2G applications is proposed. The proposed converter consists of a single-phase current-source rectifier (SCSR) and an auxiliary switching network (ASN). It offers bidirectional power flow between the battery and the grid in the buck or boost mode and expands the output voltage range, so that it can be compatible with different voltage levels. The topology structure and operating principles of the proposed converter are analyzed in detail. An indirect control algorithm is used to realize the charging and discharging of the battery. Finally, the semiconductor losses and system efficiency are analyzed. Simulation and experimental results demonstrate the validity and effectiveness of the proposed topology.

Characterization and Scalable Modeling of Power Semiconductors for Optimized Design of Traction Inverters with Si- and SiC-Devices

Silicon carbide (SiC) based power semiconductors are expected to contribute to an increase in inverter efficiency, switching frequencies, maximum permissible junction temperature, and system power density. This paper presents a comparison of silicon (Si) and SiC device technologies for the use in hybrid electric vehicle traction inverters. SiC-JFETs and SiC-MOSFETs are characterized and a scalable loss and scalable thermal modeling approach is used to find the optimum chip area for each Si or SiC traction inverter. This procedure also provides a proper technical comparison of the semiconductor technologies. The progressed simulations using standardized drive cycles and thermal-electrical coupled semiconductor models permit an inverter performance evaluation close to real load situations, leading to an improved estimation of the benefit which can be expected from systems utilizing SiC technology. This paper concludes that the SiC devices can lead to a reduction in chip area and semiconductor losses by more than 50% at the same time in hard switching applications with partial load dominated mission profiles.

Parallel single-phase ac–dc–ac shared-leg converters: Modelling, control and analysis

Reversible single-phase to single-phase dc-link converters are proposed and analyzed in this paper. They are composed of paralleled rectifiers and inverters sharing one or two legs with one or two dc-links. Modelling and a control strategy of the systems including circulating currents are developed. Compared to the conventional single-phase to single-phase dc-link converter, the proposed topologies permit to reduce: (i) the current and power ratings of the power switches, (ii) the harmonic distortion, and (iii) the dc-link capacitor losses. Additionally, the semiconductor losses of the proposed converters are smaller than that of the conventional one. Simulated and experimental results are also presented to demonstrate the viability of the proposed circuits.

New emerging voltage source converter for high‐voltage application: hybrid multilevel converter with dc side H‐bridge chain links

Hybrid multilevel converters (HMCs) are more attractive than the traditional multilevel converters, such as modular converters, because they offer all the features needed in a modern voltage source converter-based dc transmission system with reduced size and weight, at a competitive level of semiconductor losses. Therefore this study investigates the viability of a HMC that uses dc side H-bridge chain links, for high-voltage dc and flexible ac transmission systems. In addition, its operating principle, modulation and capacitor voltage balancing, and control are investigated. This study focuses on response of this HMC to ac and dc network faults, with special attention paid to device issues that may arise under extreme network faults. Therefore the HMC with dc side chain links is simulated as one station of point-to-point dc transmission system that operates in an inversion mode, with all the necessary control systems incorporated. The major results and findings of subjecting this version of the hybrid converter to ac and dc networks faults are presented and discussed.

Power losses evaluation of a bidirectional three-port DC–DC converter for hybrid electric system

Power losses of a bidirectional three-port DC–DC converter to be used in hybrid electric systems as a function of the voltage conversion ratios and the output power are evaluated in this work. An analysis and characterization of the current on the switches into the whole converter operating range are presented. This analysis allows finding the semiconductor conduction intervals, necessary to calculate the power losses. Such losses are evaluated considering both switching and conduction semiconductor losses as well as those in the transformer. The variables used in this evaluation are voltage conversion ratios and transformer parameters like leakage inductances and turns ratios. Design considerations for the high frequency transformer that allow minimizing the total power losses are proposed. Simulation results are presented in order to validate the theoretical analysis.