Search

An isolated bidirectional dc–dc converter, which combines parallel-connected two-level converters (TLCs) on the medium-voltage side and a modular multilevel converter (MMC) on the high-voltage side, namely a TLC-MMC converter, is a promising candidate for the interconnection of medium-voltage direct current (MVDC) and HVdc grids. Different from typical phase-shift controlled isolated bidirectional dc–dc converters, transformer currents of a TLC-MMC converter are regulated by the MMC in a closed-loop manner in order to achieve extremely low switching losses and low total device rating of power-semiconductor devices in TLCs. This paper addresses the dedicated operation technique and the voltage-balancing control of the TLC-MMC converter. The MMC in the TLC-MMC converter operates with six-step phase voltages and sinusoidal phase currents. Such an operation makes the MMC present remarkable differences to ac–dc applications. The direct-modulation technique, which is the most-used technique performed to MMCs in line-frequency ac–dc power conversion, has been concluded not applicable to the MMC in a TLC-MMC converter due to the significant sixth-order oscillation resulted in the dc-link voltage. Thus, the indirect-modulation technique is employed, and consequently a closed-loop voltage-balancing control is mandatory for the MMC. Possible measures for voltage balancing have been explored, and effective voltage-balancing control with a simplified implementation is proposed. Validity and effectiveness of the proposed control strategy have been verified by both simulations and experiment using a downscale prototype. [ABSTRACT FROM AUTHOR]

Electrical drive systems without dedicated speed transducer require precise information about the terminal phase voltage to achieve an acceptable low-speed performance. Recently, voltage-sensing techniques have gained in importance as they improve stability at low-speeds compared to the conventional estimation based on reference values. However, the additional sensors increase the sensitivity to sensor offset drift, which can cause self-sensing control algorithms to fail at very low speeds. In this article, a novel offset adaption method is introduced that compensates sensor offsets during operation while being fully decoupled from the control algorithm. Using a pair of phase current and voltage sensors per phase, the dead-time-related inverter voltage-distortion curve is monitored continuously. Its point-symmetrical nature is utilized for the sensor offset detection. This way, both sensor offsets can be calibrated continuously during operation and for each phase individually. This article gives a step-by-step description of the calibration algorithm and addresses implementation aspects to ensure deterministic calibration dynamics. The method is evaluated through measurements performed on a test bench for an electric-vehicle traction drive. [ABSTRACT FROM AUTHOR]

This paper presents a design concept of a sensorless control for high-speed SRM drives with small DC-link capacitor. In a cost-sensitive application, the expensive large DC-link capacitor should be replaced with a smaller DC-link capacitor for cost reduction. Using a small DC-link capacitor, the DC-link voltage extremely fluctuates during operation. The fluctuating DC-link voltage deteriorates the control stability and performance of the sensorless control. In the literature, the previous work has shown that the workable speed range of the existing sensorless control concept for high speed has been degraded to a smaller range when operating with fluctuating DC-link voltage. The proposed design concept deals with the challenge of improving the control stability of the sensorless control to operate in a wider operating speed range up to 75% (8000-35,000 rpm) under the fluctuating DC-link voltage. The control scheme is developed based on two basis sensorless control techniques, inactive phase method and current gradient method for covering the entire speed range. The proposed solutions work based on the appropriate selection of the detected signals according to the state of DC-link voltage. At the end, the functionality of the proposed technique is validated by experimental results. [ABSTRACT FROM AUTHOR]

The output voltage of power electronic converters is a very important quantity for dynamic control of power electronic systems. In electrical drives without electromechanical position or speed sensor, the terminal reference voltage is used to obtain the rotor position. Dead-time effects and semiconductor voltage drop lead to distortion in the actual output voltage and degrade the control performance when the back electromotive force magnitude is low. Thus, for stable low-speed operation, output voltage sensing becomes necessary. Due to the switching nature of power electronic systems, this is not a trivial task, especially when instantaneous measurement of the terminal voltage is required. In this paper, an instantaneous switching-period average voltage sensing technique is proposed that utilizes a combined approach of oversampling and filtering. Based on the theoretical analysis of the sampling- and filter-induced measurement distortion, a general solution for an optimal filter design is derived. The additional sensing circuit is integrated into the low-side gate driver of the converter outputs. This paper includes details about the hardware implementation and extensive verification measurements. [ABSTRACT FROM AUTHOR]

This paper presents a detailed comparison between the conventional ac collector-grid configuration and two proposed dc collector-grid configurations for large photovoltaic (PV) parks. One complete year of measured weather data at two different locations, one very sunny and the other with relatively less sun, is used for the evaluation. An already validated PV performance model is used to estimate the expected energy yield of the PV park for the given environmental and connection conditions. Loss mechanisms in different grid components in the three configurations are discussed based on the developed models. These components include the inverters, transformers, cables, and dc–dc converters. A comparison of the total losses in these components helps to identify the configuration that has the potential to operate at higher efficiency and, hence, generate maximum revenue. The economic viability metrics are calculated and a sensitivity analysis is carried out using the Monte Carlo method with a \pm \mathbf 5\% uncertainty in the considered cost and energy yield parameters. [ABSTRACT FROM PUBLISHER]

The three-phase dual-active bridge is a dc–dc converter, which is highly suitable for high-power applications. Among others, this is due to the medium-frequency transformer in the ac link, which provides galvanic isolation. The transformer is operated with a square-shaped voltage waveform. The flux density in the transformer core is piecewise linear. However, for the sake of simplicity, the magnetic flux is often assumed sinusoidal. Thereby, the actual iron losses generated in the core material are misinterpreted. This paper discusses the difference between the exact piecewise linear and the sinusoidal course in terms of iron losses. Silicon steel with a thickness of 0.18 mm is measured at a frequency of 1000 Hz, comparing the sinusoidal excitation with the actual one. The measurements are validated using the improved generalized Steinmetz equation. Finally, the transformer core losses are evaluated when the dc–dc converter is operated under load. The results are confirmed through measurement. [ABSTRACT FROM AUTHOR]

Currently, gate commutated thyristors (GCTs) are commonly used for high-power voltage-source inverters due to their superior electrical performance as compared to gate-turn-off (GTO) thyristors and high-voltage insulated-gate bipolar transistors (HV-IGBTs). However, the gate driver of GCTs is more complex than an IGBT driver and has to be attached closely to the GCT, which implies several constructional disadvantages. The internally commutated thyristor (ICT), a GCT with integrated turn-off unit, enables the use of a simplified and more reliable gate driver, maintaining the electrical advantages of the GCT. In this paper, the ICT concept is discussed in relation to standard GCTs. Additionally, an optimized gate driver for the ICT is presented. Close attention is paid to improve reliability and easy adaption to different ICT current ratings. [ABSTRACT FROM AUTHOR]

This paper presents a control scheme that allows minimization of dc-link capacitance in switched reluctance drives (SRDs). In switched reluctance motors (SRM5), the demagnetization energy of the outgoing phase causes a voltage increase in the dc-link. Therefore, large dc-link capacitors are required in conventional drives to absorb this energy to avoid over-voltages. The proposed control scheme balances the power flow between phases in such a way that the demagnetization energy of the outgoing phase does not flow back into the dc-link but directly into the incoming phase. Using this scheme, electrolytic capacitors can be replaced by film or ceramic capacitors leading to higher power density, longer lifetime, and higher reliability. [ABSTRACT FROM AUTHOR]