This study introduces the development advances and trend of power semiconductors used in hybrid and electric vehicles (HEV/EV). The status and forecast of power electronics’ requirement in HEV/EV are discussed along with a review of the automotive standard of power semiconductor devices. The advances in automotive semiconductor technologies such as Si-based insulated-gate bipolar transistor (IGBT) and freewheeling diode (FRD) and SiC-based metal oxide semiconductor field-effect transistor (MOSFET) and Schottky barrier diode are presented. The advances in automotive semiconductor packaging technologies are illustrated from three considerations, which are the low inductance in high-power density packaging, lower thermal resistance designing, and advanced packaging technologies. The challenges and development trend of more reliable power semiconductors for HEV/EV are discussed. The challenges in Si devices are focused on power density, efficiency, reliability, and packages, while the high temperature and low inductances are the main challenges for SiC devices. Lastly, the development trend is discussed in terms of four aspects, the new generation Si IGBT for HEV/EV, such as recessed-emitter-trench, reverse-conduction-IGBT, and smart IGBT; next generation SiC MOSFET; new packaging technology and material, such as planar packaging.

This work presents a hybrid 3.3 kV/450 A insulated-gate bipolar transistor (IGBT) power module, utilising the half-bridge topology. In contrast to the traditional fashion, each IGBT chip in this module is allocated with two anti-parallel silicon carbide (SiC) Schottky barrier diodes (SBDs). Each SBD describes smaller footprint than the conventionally used silicone (Si) based fast recovery diode at this voltage and current level. By adopting smaller SiC SBDs in this hybrid-style packaging structure, the SBD chip yield can be greatly improved, without any additional fabrication cost. To benchmark the advantages of this hybrid power module over the Si-based counterpart, both power modules are designed, fabricated as well as tested statically and dynamically. It is shown that while both types show similar thermal behaviour, the hybrid power module describes much lower IGBT turn-on current overshooting, diode reverse recovery loss and IGBT turn-on loss. Based on the measured results, the power losses of both modules under a three-phase inverter operation condition are calculated, showing that the hybrid module has considerably lower power losses and junction temperature rise.

Medium voltage power supplies for applications such as electrostatic precipitators are used in industrial plants to remove particles from fumes. Current solutions based on silicon devices rely on high-voltage transformers to reach the required output voltage levels. New wide band gap materials such as silicon carbide have higher electric breakdown voltage, and thus fewer devices are required in series to withstand the output voltage. Owing to the faster switching speed of silicon carbide devices further demands are put on the serialisation method. In this study, a cascaded series-connection method using only a single external gate signal is analysed in detail, guidelines to size the resistor–capacitor–diode-snubber are proposed and its applicability is experimentally demonstrated. The circuit is tested with four series-connected devices in a double pulse test at 2400 V and current levels of 250–800 mA to show the load dependence. The serialisation technique is tested in a boost converter operating in discontinuous conduction mode but is limited to 1200 V due to an oscillating state occurring after zero current crossing. Finally, the technique is tested at 2400 V and 10 kHz in a synchronous boost converter, which demonstrates the proposed design guidelines.

High-temperature electrical performances of enhancement-mode (E-mode) high electron mobility transistor with p-type Gallium Nitride (GaN) gate cap are evaluated here. The physics-based mechanisms behind the behaviours are also analysed by the simulations and analytical models. For static electrical performances, the changes of GaN bandgap and the interface states or traps are considered to be influential factors for the little variations of threshold voltage (V_T ). Meanwhile, the on-state resistance increases and trans-conductance decreases at high temperatures due to the reduction in electron mobility (µ _eff). As for blocking characteristic, high temperature-induced increase of leakage current may result from multi-reasons, such as the increase of intrinsic carrier concentration and lowering of trap barrier. In addition, a segmental method is presented to understand the gate leakage current at high temperatures. For capacitance characteristics, the increase of channel resistance makes the measured gate capacitance lower than the intrinsic value. For dynamic electrical performances, the high temperature-induced decrease of µ _eff leads to the increase of plateau voltage, bringing the decreases of total switching time and total switching energy loss, which are quite different from those of the devices with traditional metal-oxide-semiconductor structures.

To accurately estimate the switching characteristics of silicon carbide (SiC) metal–oxide–semiconductor field effect transistor, simulating with the behavioural model is a common approach. However, due to different manufacture batches, processes and application conditions, the model parameters need to be calibrated after being extracted from the datasheet. To improve the transient precision of behavioural model for SiC power MOSFET and provide technicians with more realistic and accurate simulation results, in this study, a calibration method for the behavioural model parameter is proposed by analysing the sensitivity of the parameters on the switching characteristics. The analysis shows that the sensitivity of the parameters affecting the SiC MOSFET behaviour model in sequence are C _gd, C _gs, V _th, R _g,int, g _f, L _d, L _S, C _ds and C _Dj. Using this sensitivity influence order, the corresponding model parameters can be corrected by observing the deviations between the experimental and simulation waveforms. Finally, double pulse tests were carried out and the comparison results indeed verify its accuracy under different working conditions. The proposed method is verified by Saber simulation software, and it can also be applied to other simulation software.

Owing to their excellent characteristics, silicon carbide (SiC) metal–oxide–semiconductor field-effect transistor modules are expected to have broad application prospects in various types of power conversion equipment in the future. Accurate prediction of the internal chip junction temperature of SiC module is of great value for the usage of the modules in the power conversion equipment. In this study, taking the electrothermal coupling process, and the thermal characteristics of the chips into consideration, the power loss model under unipolar and bipolar sinusoidal pulse-width modulation control was developed. A star-shaped equivalent thermal network model based on virtual temperature was established. Moreover, an analytical junction temperature predicting model was proposed. To obtain the temperature dependence of static and dynamic parameters for the models, power device analyser was used, and the double pulse test bench was established. To validate the effectiveness and accuracy of the above models, the finite element method was used to calculate the junction temperature of a commercial 1200 V/300 A full SiC module under different working conditions. The results show that the relative error is very small and the proposed model is effective.

SiC MOSFETs (silicon carbide metal-oxide semiconductor field-effect transistors) are replacing Si insulated gate bipolar transistors in many power conversion applications due to their superior performance. However, ruggedness and reliability of SiC MOSFETs are still big concern for their widespread applications in the market, especially in safety-critical applications. The objective of this study is to provide a comprehensive picture on the ruggedness and reliability of commercial SiC MOSFETs, discover their failure or degradation mechanism, and propose some possible mitigation methods through both literature survey and in-depth analysis. The ruggedness of SiC MOSFETs discussed here includes short-circuit (SC) ruggedness, avalanche ruggedness, and their failure mechanism. The reliability issues include gate oxide reliability, degradation under high-temperature bias stress, repetitive SC stress, avalanche stress, power cycling stress, body diode's surge current stress, and their degradation mechanism. Furthermore, this study discusses methods and solutions to improve their ruggedness and reliability.

High-voltage enhancement-mode (E-mode) gallium nitride (GaN) high electron mobility transistor (HEMT) is a superior candidate to enable higher efficiency and higher power density when compared with silicon power devices in power converter applications. However, the dynamic problem affects the conduction loss of the converter and remains one of the major issues that must be resolved. In this study, a comprehensive experimental evaluation and analysis method of the temperature-dependent dynamic of GaN HEMT in a circuit level is proposed. A commercial E-mode GaN HEMT (GS66508T) is used as a sample to study the temperature-dependent dynamic using a double-pulse-tester. The temperature-dependent dynamic under different DC-link voltages and different load currents are studied and the results show a non-monotonic temperature-dependence of the dynamic . It is concluded that the temperature dependence of the buffer-induced trapping and de-trapping effect, and the temperature dependence of electron mobility together influence the dynamic of E-mode GaN HEMT device during operation. This finding is important since in converter applications the devices are typically operating at elevated temperatures. The proposed comprehensive experimental method can be used to estimate and analyse the dynamic characteristics of other GaN devices.

Operating silicon carbide (SiC) metal–oxide–semiconductor field-effect transistor (MOSFET) at its rated switching speed may not be always feasible due to excessive voltage and current overshoot and ringing caused by layout parasitic inductance and load parasitic capacitance. This study proposes a low-cost analogue active gate driver technique for switching SiC MOSFET in the presence of moderate amount of layout parasitic inductance (<200 nH) and load parasitic capacitance (<300 pF). In this study, a based closed-loop active gate driver circuit is implemented using low-cost signal level transistors. The present work explains the working of the gate driver during turn-on and turn-off switching transients. A detailed design methodology for the gate driver is presented using its high-frequency model. The proposed active gate driver (AGD) has been verified in hardware platform using a double pulse test setup and in a boost converter test setup. Cree make 1200 V, 36 A SiC MOSFET (C2M0080120D) is used for evaluating the proposed active gate driver. The proposed circuit has sufficient operating bandwidth to drive SiC MOSFET and it is realised with low-cost transistors.

Industrial processes which use induction and dielectric heating are still relying on resonant converters based on vacuum tubes. New emerging medium-voltage silicon carbide (SiC) semiconductor power devices have a potential to replace vacuum tubes and allow for more efficient and compact converters in the high-frequency range. High-voltage packages have been proposed in the literature that are suitable for the 10 kV SiC metal–oxide–semiconductor field-effect transistors (MOSFETs), and its fast voltage switching capabilities in hard-switched applications have been demonstrated. However, no packaging is presented which allows the high-frequency operation of a 10 kV SiC MOSFET die. This study proposes the design of a power module for MHz resonant operation of a 10 kV SiC MOSFET. At high switching frequencies, the gate losses become substantial, thus the gate driver is included inside the power module package to ensure a low inductive and high thermally conductive design as seen from the gate driver. The inductance of the proposed power module layout structure is evaluated using ANSYS Q3D Extractor. The thermal performance of the integrated gate-driver circuitry is experimentally verified. Finally, the resonant operation of a medium-voltage SiC MOSFET power module is demonstrated experimentally at 1 MHz.

This study deals with the design, implementation and experimental performance of all silicon carbide pulse width modulation (SiC PWM) rectifier-based off-board ultrafast chargers (UFCs) for lithium titanate batteries of heavy electric vehicles. Different UFC configurations are proposed, depending upon the nominal battery voltage (400–1000 V DC) and charge capacity. Operating principles and control of UFC at unity and leading power factors are assessed and corresponding operating modes of PWM rectifier are discussed. The combined effect of reverse conduction characteristic of SiC power metal oxide semiconductor field effect transistor (MOSFET) and built-in SiC Schottky diode is taken into account in all the analyses carried out. The operating performance of the developed UFC, such as the switching characteristics of SiC power MOSFET modules, efficiency, input current total demand distortion (ITDD), and thermal limitations of the SiC PWM rectifier have been assessed for various charge voltages and charge capacities, both by computer simulations and laboratory tests. Power circuit layout considerations of the proposed system are also given in this study. Excellent performance results for 10 kHz switching frequency are obtained from the developed 200 kW UFC, with operating efficiencies higher than 98.5% for all charging rates up to five times the battery capacity, and ITDDs <2.2% for the whole operating range.

A two-mode resonant DC–DC converter with auxiliary switch (named as sLLC-C) has been presented. Two simple structural circuits (namely LLC and LLCC) are integrated into one topology with only two magnetic cores, through the topological transformation method. The proposed converter successfully inherits the advantages of the above two circuits, while avoiding their drawbacks as well. At the rating condition, it functions as an LLC circuit, and thus high conversion efficiency is easily obtained. The LLCC working mode is called automatically once the input voltage increases. In such a way, within a limited frequency range, the voltage gain range is effectively extended. A detailed parameter design is given to help achieve desirable performances. Finally, experiments are carried out to validate the feasibility and correctness of the proposed sLLC-C converter. The output voltage is maintained at 400 V, even if the input voltage varies in the range of 150–400 V. Besides, the preferable soft-switching characteristics are ensured for both the power switches and secondary-side diodes. The maximum efficiency of 97.3% is achieved. Thus, wide voltage gain range and high conversion efficiency are acquired simultaneously. These performances make the proposed converter well appropriate for the fuel cell generation applications.

This study presents a novel family of dual-transformer multi-resonant DC–DC converters (DTMRCs) with a resonant zero point (RZP). The limitations of the conventional LLC resonant converter and its fundamental harmonic approximation (FHA) method are firstly discussed. On this basis, a unique RZP is combined with the dual-transformer architecture. As a result, both of the output voltage regulation capability and the over-current protection feature are promoted. Then, an extended describing modelling method is presented to improve the calculative accuracy of the FHA method. Moreover, by means of the appropriate placement of all three resonant frequencies, the fundamental and the third-order harmonic components are employed to transmit active power to the load simultaneously. The zero-voltage soft-switching characteristic is acquired for all the primary-side switches. Meanwhile, the preferable zero-current soft-switching feature is obtained for diodes both at the turn-on and turn-off moments. Therefore, the switching losses are significantly reduced for the proposed converters. Finally, with the employed SiC metal–oxide–semiconductor field-effect transistors, a 500-W half-bridge DTMRC prototype is built as an example, based on the P-type multi-resonant network. The experimental results demonstrate the validity and correctness of the theoretical analyses.

For high-frequency high-current low-voltage applications, planar matrix transformer for unregulated LLC converter has been adopted to achieve high power density and high efficiency. Compared to unregulated LLC converter, the phase shift between the primary current and the secondary current of the matrix transformer is larger in regulated LLC converter as a result of lower magnetising inductance. Thus, the conventional winding loss model is not applicable to the matrix transformer for regulated LLC converter. Aiming at this problem, an accurately winding dc resistance model and an analytic winding ac resistance model are proposed in this study to calculate the winding loss of the matrix transformer. In addition, lower magnetising inductance also causes more serious fringing field near the air gap. Then, a calculation method to model the winding loss due to the fringing field is investigated. With the proposed model, design and optimisation of the matrix transformer for regulated LLC converter are described in detail subsequently. The proposed model is verified by a Gallium Nitride based 400 W LLC converter prototype with a power density of 443 W/in³ and a peak efficiency of 96.1%.

This paper explores performance enhancement of the common ground dynamic dc-link (CGDL) inverter for single phase photovoltaic (PV) applications by a combination of gallium nitride (GaN) devices, split phase topology, coupled inductors, and zero voltage transition (ZVT) scheme. The CGDL inverter has the inherent advantage of minimised dc-link capacitance and negligible leakage current due to the common ground configuration, but its reported efficiency was usually lower because of the higher dc-link voltage used for the reduction of decoupling capacitance to a great extent. To solve the efficiency problem, in this study, a soft switching circuit is proposed for the first stage, while a coupled inductor integrated magnetics is incorporated in the second stage to reduce inductor loss, volume, and cost. Both of these topological improvements combined with the use of GaN devices facilitate in achieving high efficiency without compromising converter power density. Extensive experimental results are provided from a GaN based 1 kVA hardware prototype to demonstrate the superior performance of the CGDL inverter attaining a peak efficiency of 98.7% and a California Energy Commission efficiency of 98.5% at 75/50 kHz switching frequency.

A power factor corrected single switch tapped-inductor buck–boost converter is proposed to achieve improved efficiency in extreme step-up and step-down conditions. This new and modified topology is having diode-tapped and switch-tapped configurations for step-down and step-up modes of operation, respectively. The proposed modified buck–boost converter is designed to operate in critical conduction mode (CrCM) to achieve quasi-resonant operation. Both the buck and boost functionalities of the buck–boost converter are analysed using state space analysis with closed loop stability under varying input and output conditions. Several advantages and necessity of tapped versions are discussed and implemented under wide input and extreme output (WIEO) conditions. Two different prototypes are developed at extreme buck and boost levels to establish the benefits of the proposed arrangements. The efficiency improvement of the proposed converter is compared with the conventional buck–boost converter. The achieved efficiency of diode-tapped buck mode and switch-tapped boost mode is found 95% and total harmonic distortion is observed only 4% at 230 V AC mains for a prototype of 150 W, which is much improved results as compared to the previous research work on the tapped-inductor converter.

In this study, a new three-phase current-fed push–pull DC–DC converter with high-gain capability was proposed. The proposed circuit has the advantages of three-phase converters and the push–pull converter circuit, with small volume and a simplified gate drive circuit with switches connected to the same reference. The converter was analysed for operation in continuous and discontinuous conduction modes. The analysis includes a description of the topological stage, theoretical waveforms, deduction of the voltage gains and the expressions for its practical design. To validate the theoretical analysis, a 700 W prototype was built and tested with an input voltage of 40 V, an output voltage of 400 V and a switching frequency of 40 kHz. Despite using a unity transformer turns ratio, the prototype had an experimental gain of ∼10, which makes the proposed converter suitable for applications as renewable energy sources systems.

It is a challenging task to design controller for the digitally controlled LCL-filtered grid-connected inverter to achieve high-quality grid current, avoid undesired resonance, and offer robust performance against system parameters uncertainties. To address the problem, in this study, the linear quadratic regulator (LQR) for a digitally controlled grid-connected inverter is proposed. Discrete-time state-space model of the inverter is established by discretising the system continuous state-space equations and taking into account one sampling period delay. Then, a systematic design procedure for selecting the state weighting matrix Q is presented, so that the poles of the closed-loop system are assigned and the desired performances can be achieved. Moreover, to further attenuate the grid voltage background harmonics, the selective harmonic controllers are designed based on the LQR. Compared with the conventional dual current loop control strategy, the proposed LQR controller can achieve low harmonic distortion, fast dynamic response, and good robustness against system parameters deviations. Finally, time-domain simulations are carried out in MATLAB/SIMULINK software and experimental tests are performed on the experimental setup. The results have demonstrated the effectiveness of the proposed LQR method.

To eliminate harmonic components in estimated speed and position information generated by inverter non-linearities for the interior permanent-magnet synchronous motor (IPMSM) sensorless control, a second-order lead compensator-based quadrature phase-locked loop (SOLC-Q-PLL) is proposed for the sliding mode observer. An adaptive super-twisting sliding mode observer is taken to estimate back electromotive forces (EMFs) for IPMSM. Harmonic components in estimated position information generated by inverter non-linearities are analysed. Being dependent on the corresponding analysis, the SOLC-Q-PLL is proposed to eliminate sixth position estimation errors only by introducing a second-order lead compensator without causing phase delay for the phase-locked loop system. The introduction of the second-order lead compensator does not influence the bandwidth of the phase-locked loop system so that a fast transient response can be achieved, and this algorithm is easy to implement digitally due to its simple structure. Eventually, a series of experiments are conducted to validate the capability of the proposed algorithm.

This work proposes a soft-switched quadratic quadrupler boost converter with high voltage step up gain and low voltage stress on devices. With this converter topology, high voltage gain is achieved at low duty ratio operation of MOSFETs and few number of turns in coupled inductor. Operation at low duty ratio keeps boost converter gain in linear region and smaller number of turns in coupled inductor reduces the lossy part of the winding. Further, the magnetising inductance of coupled inductor is utilised to achieve zero voltage switching of MOSFETs to minimise the switching loss. In addition, secondary side diodes are operating under zero current switching conditions using resonance between leakage inductance and the capacitors of voltage quadrupler circuit. This greatly reduces the reverse recovery losses of the secondary side diodes. A detailed analysis of converter dynamics is discussed in one of the subsections to find out small-signal transfer functions with respect to variations in input voltage and duty ratio variation of MOSFETs. A 250 W prototype of the proposed converter is built and tested in laboratory with a maximum efficiency of ∼93% at 175 W output power.

A hybrid full-bridge converter that consists of a secondary-side phase-shift full-bridge converter (SPS-FBC) and a full-bridge LLC converter is proposed in this study. The wide zero-voltage switching (ZVS) range of the primary switches can be achieved and the series inductor of the SPS-FBC converter does not need to be specially designed to help achieving the soft switching operation with a set of shared high frequency inverter circuits. The dual outputs of the proposed hybrid converter are connected in series, reducing the voltage stress of the power device in the secondary side. The whole dc-output voltage can be regulated by primary-side phase-shift and secondary-side phase-shift dual-mode control scheme within the desired voltage range, by which the two output voltages are regulated independently and free from cross-regulation. Therefore, the proposed converter is suitable for high voltage, high power and wide output voltage range applications. The working principle and steady-state characteristics of the proposed converter are presented and verified in experiment of a 3 kW prototype. The experimental results show that the proposed converter maintains high efficiency throughout the voltage range, and the peak efficiency is 97.43%.

The virtual synchronous generator (VSG) is beneficial to reduce the fluctuation of the output power of the renewable energy sources, but its output power characteristics are sensitive to the disturbance of the grid frequency. Grid frequency disturbance causes the oscillation of the output power of the VSG-controlled grid-tied converter. In order to solve this problem, a virtual inertial control method for renewable energy sources without additional energy storage is proposed. In the view of renewable power generation, the renewable energy source behaves as a VSG, which minimises the impact of power generation disturbances on frequency characteristics. In the view of grid frequency disturbance, the renewable energy source behaves as a droop-controlled grid-tied converter, which adapts to the grid frequency to avoid the power oscillation of grid-tied renewable power source caused by the grid frequency disturbance. Grid frequency is introduced into the closed loop control through feedforward techniques, and the frequency extraction method is implemented based on a frequency-locked loop (FLL) modified by a moving average filter. Based on the small-signal model of the proposed VSG method, the influence of the equivalent FLL gain and time constant is analysed. The experimental results verify the performance improvement of the modified VSG method.