Wei Wang

Silicon-carbide (SiC) solid-state circuit breakers (SSCBs) are essential components of flexible direct current (DC)-traction systems used in urban-rail transportation because they facilitate rapid and arc-free fault isolation. However, SSCBs using single SiC devices are inadequate for medium-voltage, high-current applications. Therefore, a novel SSCB topology and single-gate drive circuit were proposed, using a series-parallel configuration of normally-on SiC junction field-effect transistors (JFETs). The main switch adopted a three-series and two-parallel connection using six 1.2 kV/63 A normally-on SiC JFETs. The proposed single-gate drive circuit achieved both static and dynamic balancing of voltage and current, among all the SiC JFETs—during short-circuit fault isolation and recovery—and required only a few passive components. The SSCB operating principles during short-circuit and recovery processes were analyzed. A key-parameter optimization method is presented in this paper. Experiments on a 1.5 kV/126 A SSCB prototype demonstrated the performance of the circuit configuration presented. Results showed that the SSCB interrupted a 500 A short-circuit fault (fault resistor was 3 Ω) within ~16 μs, maintaining superior voltage and current uniformity during the entire switching cycle.

Solid-state circuit breakers (SSCBs) enable rapid and reliable isolation of short-circuit faults in photovoltaic (PV) storage and charging dc microgrids. However, the main switch typically experiences high current and voltage stress during fault isolation, thus challenging the safe operation of the SSCB. Hence, a novel SSCB topology based on a normally-on silicon carbide (SiC) junction field-effect transistor (JFET) is proposed, which can achieve fully zero-current turn-off for both the main and auxiliary switches. During a short-circuit fault, an $LC$ resonant circuit injects a reverse pulse current into the SiC JFET main switch to achieve zero-current turn-off, while a metal-oxide varistor (MOV) branch absorbs the fault current flowing through the SiC metal-oxide-semiconductor field-effect transistor (MOSFET) auxiliary switch, thereby facilitating the zero-current turn-off of the SiC MOSFET. Subsequently, the operating characteristics and design considerations for key components, including the inductor, capacitor, and MOV, are analyzed. Finally, a 375 V/63 A SSCB prototype is constructed and validated. Experimental results confirm that the proposed topology achieves zero-current turn-off for all the switches when interrupting a 260 A fault current, thereby significantly enhancing the interrupting performance and energy efficiency of the SSCB in PV storage and charging dc microgrids.

The on-state voltage of silicon carbide (SiC) MOSFETs (V_dson) serves as a critical parameter for reliability assessment and performance evaluation, demanding precise online measurement. However, existing solutions face three key challenges in high-frequency applications: (1) uncompensated voltage-drop errors in diode-based circuits; (2) sluggish dynamics of MOSFET-based designs due to parasitic capacitance; (3) dependency on control signals for switch identification in converter-level monitoring. This work proposes a simplified converter-level on-state voltage monitoring circuit (OVMC) with dual-path architecture, enabling control-free V_dson measurement in LLC converters operating at 100 kHz. A phase-sensitive digital compensation strategy further mitigates errors from diode voltage drops (V_D) and parasitic inductance (L_p). Experimental validation on an LLC platform demonstrates D compensation in V_dson extraction, with a circuit response time <400 ns.

The subsea direct current (dc) converter is prone to component failures, particularly in switching devices and input capacitors, primarily due to its installation within a tightly sealed, small-volume metal enclosure, which significantly impairs heat dissipation. To enhance the reliability of subsea input-series output-parallel (ISOP) dc converters, this article proposes a multilevel synergistic fault-tolerant strategy that takes into account the challenges of large input voltage fluctuations and limited electrical monitoring. Unlike conventional fixed-threshold methods, the proposed scheme leverages the dynamic voltage–current time-domain signatures of submodules during incipient faults for real-time monitoring and fault localization, thereby maintaining robustness against input voltage variations. Furthermore, a redundant bypass circuit based on current-limiting resistor multiplexing is incorporated to improve the traditional topology. When combined with a smooth redundancy switching mechanism, this approach effectively mitigates transient electrical stresses on healthy modules during fault transitions. These enhancements collectively ensure the operational safety of both the subsea converter and associated high-precision deep-sea equipment. Finally, a comprehensive fault-tolerant control scheme is developed to guarantee the reliable operation of the subsea dc power supply system following fault occurrences.

Modular cascaded LLC converters are widely used in high-power scenarios of the DC power system with wide voltage gain, not limited by the voltage withstand level of existing equipment. However, the cascaded structure brings a larger number of capacitors and may increase the failure risk of the entire system if one of them fail. However, monitoring these capacitors require complex and multiple monitoring circuits if the conventional solutions are applied. Hence, a simple way to realize the health monitoring of all of them is preferred and is rarely discussed for the modular cascaded LLC converter. In this paper, an online monitoring scheme for multiple capacitors based on only one-time transient large-signal trajectory is proposed for an input-series-output-parallel type LLC cascade converter. Instead of adding additional dedicated sensors, the scheme utilizes the existing sensors and extracts the input capacitance, resonance capacitance and output capacitance from the large-signal trajectory simultaneously, without affecting the normal operation of converter. Finally, experiments are conducted to verify the effectiveness and accuracy of the method for capacitance monitoring in an LLC converter platform.

More and more distribution generators (DGs) are integrated into renewable energy systems, which increases the transient instability risk of systems. The Takagi-Sugeno (TS) fuzzy multimodeling method is usually applied in transient stability analysis of power electronic systems. However, the traditional TS fuzzy multimodeling method is based on the linear matrix inequalities (LMIs) calculations and attraction domains search with the fixed search directions and step size, which severely increases conservation and calculation burden while evaluating the stability of high-order systems. To fill this gap, a large signal stability analysis method based on attraction domains directionless search is proposed in this paper. Firstly, the drawbacks of the traditional TS fuzzy multimodeling method is analyzed. Secondly, an active attraction-domain-search-based (AADSB) Lyapunov function construction method is proposed. The attraction domain index (ADI) is defined in this paper. The proposed method constructs Lyapunov function by searching the maximum ADI of studied systems. The position of the operating point which is corresponded to the maximum ADI is used to calculate the candidate energy function of studied systems. Thirdly, simulations and experiments are proposed to verified the effectiveness of the proposed method. Compared with the traditional TS fuzzy multimodeling method, the proposed method is less conservative and has less parameter selection restrictions.

Due to the high incidence of faults in submarine cables, the rapid and accurate fault line detection is crucial for ensuring the safety and reliable operation of submarine MVDC power systems. In this paper, considering the fault response of shore station MMC, the transient and steady-state fault characteristics of meshed submarine MVDC power systems are analyzed. It reveals the differences in the direction and amplitude of high frequency superimposed currents (HFSCs) and low frequency superimposed currents (LFSCs) at both ends of the fault and non-fault lines. Then, by extracting transient energy from HFSC and monitoring changing trend of LFSC, signed binary state direction features of full frequency current are derived. These features are used to construct the protection criteria, which are robust against the change of fault scenario, fault starting speed and sampling frequency. Localized feature processing renders the protection accuracy immune to communication delay and synchronization error. Furthermore, the low bandwidth fault data exchange demand between line terminals is ensured by combining the fast submarine cabled communication and flexible space-ground-sea integrated network. Finally, simulation results on the modified North East Pacific Time-Integrated Undersea Networked Experiments (NEPTUNE) power system validate the effectiveness of the proposed scheme under various fault scenarios.

Due to the high fault occurrence probability of submarine cable caused by extreme working environment, accurate fault detection and location is very essential for submarine DC power systems. While electrical faults are usually accompanied by the failures of optical fiber communication, resulting in the poor reliability of the existing methods based on cabled communication. Therefore, we propose a communication-free fault location method using multiple incipient current features. First, the fault modeling and current loop decoupling of submarine DC power systems are carried out to obtain the time domain description of fault line current. According to the varying trends and rates of incipient fault current with fault distance and fault resistance, multiple time domain features are utilized to reflect the fault distance meanwhile weaken the adversity of transition resistance. Then, the segmented mapping model between selected features and fault distance are established by using least squares support vector machine regression (LSSVR) algorithm. Via the integrated fault features and segmented location models, the proposed method can cover the fault location of whole line, and offers the advantages of fast speed and high accuracy. Finally, PSCAD simulation and hardware-in-loop (HIL) experiment results validate the effectiveness of the proposed method under various fault conditions.

Solid-state circuit breaker (SSCB) has the advantages of fast switching response, no arc, and flexible control in dc fault clearance. The bidirectional power flow of dc microgrid requires SSCBs based on power devices with unidirectional turn-OFF capability to reliably clear faults at either end. Thus, this article proposes a self-powered bidirectional SSCB (BSSCB) without an external auxiliary power supply. First, a self-powered BSSCB topology of silicon carbide junction field-effect transistors (SiC JFETs) is proposed. Then, the turn-OFF process of the proposed BSSCB is analyzed in detail, where the effects of self-powered capacitors and voltage-clamping metal oxide varistors (MOVs) are considered. It reveals the influence rules and parameter design basis of these components on turn-OFF time and overshoot voltage. Finally, a BSSCB prototype is constructed and tested. The experimental results demonstrate that the proposed BSSCB can rapidly isolate short-circuit faults. Additionally, it can maintain a safe and stable self-power supply whether the short-circuit fault occurs on any side of the BSSCB.

LLC converters are widely used in high-power, high-power-density scenarios due to their ability to achieve high-frequency operation and galvanic isolation. However, the ac resonant capacitor and output dc capacitor are prone to aging and failure under high-frequency charging and discharging conditions. This article proposes an online condition monitoring scheme for the multicapacitors in the LLC converter based on an active control transient. The motivation behind the active control is to create a ms-level discharge processes during the converter operation while maintaining the output voltage fluctuations within an acceptable level. During this discharging transient, the output voltage, output current, and resonant current of the converter are recorded only. Subsequently, a capacitance estimation algorithm based on a back propagation neural network is proposed and embedded in the microcontroller. Finally, experiments verify the effectiveness of the method for capacitance monitoring in the field, and its estimation error is less than 3% for the resonant capacitor and less than 1.5% for the output capacitor.