ABSTRACT With voltage source converters (VSCs) being widely integrated into the power‐electronised power system, their quasi‐steady‐state instantaneous power flows (QSS‐IPFs) under harmonic emissions and resonance risks have become an important power quality concern. However, research on QSS‐IPFs in VSC grid‐connected systems still lacks an integrated analytical path linking indicator definition, mechanism characterisation and device‐level optimisation. Therefore, this article conducts research along this path. A QSS‐IPF indicator framework and its hardware‐in‐the‐loop self‐consistency validation are introduced, with the indicators partitioned according to the system resonance characteristics and topological property. The generation and conversion mechanisms of constant, pulsating, and oscillatory components of QSS‐IPFs within the VSC are investigated, with the geometric algebra channel‐energy formalism applied to the VSC switching‐function model to provide the analytical basis. Based on the instantaneous power oscillation energy (IPOE) and its quantified transmission characteristics of QSS‐IPFs, a PSO‐based LCL filter optimisation method with its optimisation objective built on the IPOE is proposed. Together with the QSS‐IPF indicator block derived from the existing harmonic coupled source model, this method supports the optimisation analysis of grid‐connected QSS‐IPF pollution.
ABSTRACT When renewable energy is exported through an islanded modular‐multilevel‐converter‐based high‐voltage direct‐current (MMC‐HVDC) system, the sending‐end converter station usually operates in voltage‐frequency (VF) control mode to support the alternating‐current (ac) voltage and frequency. The renewable energy side behaves as a current source. This source characteristic mismatch makes the power absorbed by the VF‐controlled converter difficult to regulate directly. Under converter blocking faults, the power balance may be rapidly disturbed. The resulting power imbalance can cause dc overvoltage and arm overcurrent within several milliseconds to tens of milliseconds. However, MMC‐HVDC converters have limited overvoltage and overcurrent withstand capability, and conventional security control devices are usually too slow for this transient process. To address this problem, this paper studies blocking faults at both sending‐end and receiving‐end converter stations. The mechanism of power uncontrollability in the VF‐controlled converter is analyzed. Operating power‐margin design principles are proposed for safe dc transmission operation. The influence of renewable energy integration topology is also investigated. A coordinated direct‐power control strategy among multiple converter stations is further developed to rapidly redistribute unbalanced power. An electromagnetic transient model of a practical MMC‐HVDC grid project is used for verification. The results show that the proposed method can improve post‐fault power coordination and suppress dc overvoltage and arm overcurrent within the designed operating margins.
ABSTRACT This study investigates low‐frequency oscillations (LFOs) in traction power supply networks coupled with electric locomotives. A state‐space averaged model is developed to characterise the system's nonlinear, non‐autonomous, and time‐varying properties. To address the limitations of traditional small‐signal analysis, the HB‐Floquet method, integrating Harmonic Balance and Floquet theory, is proposed to quantitatively assess periodic solutions, stability boundaries, and bifurcation behaviours under high‐power, strongly nonlinear conditions. Crucially, motivated by the low‐frequency oscillation phenomena initially captured during high‐voltage field measurements, the proposed theoretical framework accurately replicates these practical occurrences. The theoretically predicted LFO frequencies and stability margins are quantitatively benchmarked against field measurements. Theoretical analysis and simulations reveal that increasing the voltage‐loop integral gain K vI can induce LFO through a Hopf bifurcation. The highly consistent quantitative trends observed among the theoretical averaged model, time‐domain simulations, and field measurements verify the effectiveness of the proposed method. Overall, this work provides a validated theoretical framework for the parameter design, optimisation and operational stability of complex traction power networks.
ABSTRACT This paper presents a comprehensive modulation‐based performance enhancement of a quadruple‐boost nine‐level switched‐capacitor inverter (QB9LSCI) for improved power quality, efficiency and practical implementation. Two level‐shifted carrier waveforms, namely WM‐type and UM‐type carriers, are proposed for high‐frequency PWM operation and compared with conventional triangular and full‐parabolic carriers. The modulation methods are evaluated in terms of voltage and current total harmonic distortion (THD), capacitor‐voltage ripple, semiconductor loss distribution, efficiency and dynamic response. The proposed WM‐type carrier provides the best performance among the carrier‐based methods, reducing the output‐voltage THD to 5.71% and current THD to 0.56%, while achieving lower capacitor‐voltage ripple and higher efficiency. For low‐switching‐frequency operation, selective harmonic elimination (SHE) and selective harmonic mitigation (SHM) are implemented using bee algorithm‐optimised switching angles. The bee algorithm converges within approximately 30 iterations, compared with 150–200 iterations required by PSO, DE and GA under identical optimisation settings. The SHM method further reduces the output‐voltage THD from 10.63% (SHE) to 6.63% by mitigating a wider harmonic range within prescribed limits. Loss and efficiency analyses show that SHE and SHM significantly reduce switching losses, making them suitable for high‐power steady‐state operation, whereas WM‐carrier PWM provides superior controllability and dynamic response for medium‐power applications. The proposed modulation strategies are validated through MATLAB/Simulink, PLECS loss analysis and experimental results from a 500–1000 W hardware prototype. The results demonstrate that the proposed QB9LSCI with the proposed modulation strategies offers an effective trade‐off between harmonic performance, efficiency, implementation simplicity and application flexibility.
ABSTRACT Advances in wide‐bandgap (WBG) semiconductors now enable inverter designs capable of operating at higher speeds, higher voltages, and significantly greater power levels. Operating at high fundamental and switching frequencies enables improved power density, reduced current and voltage distortion, and enhanced overall drive efficiency. However, their practical extension to low‐voltage megawatt‐class systems is constrained by device paralleling, semiconductor losses, thermal management, current distortion, and implementation complexity. This paper presents a PLECS‐based electrothermal feasibility framework for extracting feasible operating boundaries of MW‐class SiC inverters for high‐speed drives. The main objective is to identify the maximum feasible switching frequency and engineering applicability limits of different topology, device, DC‐link, and modulation‐technique combinations under simultaneous constraints of efficiency, semiconductor junction temperature, and output‐current total harmonic distortion (THD). Two‐level, T‐type neutral‐point‐clamped (T‐NPC), and active neutral‐point‐clamped (ANPC) inverter topologies are evaluated using commercially available SiC discrete MOSFETs and power modules at 800 V and 1500 V DC‐link levels. The study evaluates switching and conduction losses, total harmonic distortion (THD), efficiency, and junction temperature across varying frequencies and modulation strategies, including space vector pulse‐width modulation (SVPWM) and discontinuous pulse‐width modulation (DPWM) variants. Suitable feasible solutions achieving efficiency, THD , and balanced cost–thermal trade‐offs are presented. A 25 kW SiC inverter prototype is used to validate the PLECS electrothermal model and support the predicted loss, efficiency, and temperature trends.
ABSTRACT Three‐phase filter capacitors are key components in urban rail vehicle auxiliary converters. They filter the output ripple voltage and ensure a stable power supply for onboard equipment. Their performance degradation poses significant safety risks. This paper proposes an online capacitance estimation method for condition monitoring. The method uses existing voltage and current sensors of the converter without adding hardware costs. To address the challenges of the difficult separate estimation of three‐phase capacitance values and high harmonic content in the data, an improved estimation model is developed. This model suppresses switching harmonics while reducing the impact of sensor DC bias. An improved Whale Optimisation Algorithm with a deterministic fitness‐level‐based update mechanism is adopted to achieve accurate capacitance estimation through search. Furthermore, comprehensive robustness analysis reveals that a minimum sensor accuracy of Class 0.5 is required to provide sufficient diagnostic margin for a typical 10% early warning threshold. Experimental validation on a subway auxiliary converter shows that the estimation error of this method is below 3.5% compared to measured values. The proposed method demonstrates better stability than other algorithms and effectively reduces misdiagnosis risks caused by measurement noise and harmonic distortion.
ABSTRACT Online stability margin monitoring is critical for ensuring the reliable operation of grid‐connected inverters under varying grid conditions. However, existing impedance‐based methods are constrained by measurement accuracy and struggle to balance flexible margin regulation against mathematical simplicity. To address these issues, this paper proposes a robust online monitoring framework. First, a measurement scheme based on the injection‐to‐grid current ratio is introduced. Unlike approaches relying on impedance ratios, this method uses the dominant perturbation energy to improve the signal‐to‐noise ratio (SNR) and eliminate the dependency on high‐bandwidth voltage sensors. Second, a Rational Algebraic Stability Criterion (RASC) is developed. It constructs a polynomial boundary that decouples gain margin and phase margin constraints, offering high design flexibility without using complex transcendental functions. Finally, based on the RASC, a normalised stability margin index (NSMI) is derived to transform geometric constraints into a continuous scalar indicator for online stability margin monitoring. Experimental results validate the robustness of the proposed scheme under various grid conditions.
ABSTRACT To enhance the parameter robustness of model predictive control for grid‐tied inverters in offshore wind power grid‐tied systems, an advanced inductance identification method based on a minimum‐order state observer is proposed. First, an improved grid voltage observer with merely one gain parameter is designed. Besides, an inductance identification approach based on a grid voltage observer is proposed. Second, a thorough analysis of the observer characteristics is carried out. It is discovered that by adjusting the observer gain, a trade‐off can be achieved: a larger gain accelerates the dynamic convergence speed, while a smaller one fortifies its robustness against grid frequency deviations. This offers valuable guidance for parameter design. Finally, comparative experimental studies are conducted, which validate the effectiveness of the proposed method in improving the robustness of model predictive control parameters.
Integration of the grid-forming inverter is critical for the stability of future power grids, but their stability in the presence of large disturbance remains a challenge. Most of the existing stability analyses neglect or simplify the DC-side dynamics, even though this may be one of the main sources of instability of the whole system. Given this gap, the understanding of the limits of the system can often be incomplete and may obscure some important failure risks. This paper addresses the transient stability of GFM inverters deeply, considering the contribution of the DC side to the system. The authors presented a comprehensive model that integrates electrical AC side behaviour with the physical dynamics of an energy storage system and a photovoltaic source. This new framework addresses these previously overlooked issues and reveals how instability problems can originate from the DC side. Results from this study show that reducing the capacity of the ESS actually enhances stability, owing to the fact that it helps to dampen the oscillations resulting from an aggressive control strategy. It also points out one risk: high PV generation could actually threaten stability. A large surplus of DC power during an AC fault could result in overvoltage on the DC link, possibly leading to loss of synchrony and inability to recover by the system. Thus, this research provides a more precise way to assess the stability of GFM inverters. The findings are supported by detailed simulations and are further validated by experiments showing the existence of such DC-driven instability risks.
This paper proposes a dual-input battery charger (DIBC) with grid power and solar energy. The proposed DIBC consists of a boost-type power factor correction (PFC) circuit, a high step-down DC converter (HSDDC), and a buck switch. The grid charges the battery through the boost-type PFC circuit combined with the HSDDC, while the solar module charges the battery via the buck switch integrated with part of the HSDDC. To simplify the power circuit of DIBC, the boost-type PFC circuit and the HSDDC share a common power switch. Additionally, the buck switch shares components of the HSDDC to implement a buck DC converter. The PFC circuit operates in discontinuous conduction mode. To reduce harmonic distortion in the AC input current, the power switches of the HSDDC are modulated using a pair of signals consisting of a DC component and a second-order harmonic component. Furthermore, the second-order harmonic components of the two modulation signals are out of phase to reduce the ripple in the DC output voltage. The control of the buck switch incorporates maximum power point tracking to maximise the output power of the solar module. A 200 W hardware prototype is implemented to verify the performance of the proposed DIBC.
ABSTRACT Permanent magnet synchronous motors (PMSMs) are widely used in high‐performance servo systems and electric transportation, placing higher demands on the dynamic response, steady‐state accuracy and robustness of speed control systems. Model predictive direct speed control (MPDSC), which directly optimizes the rotor speed based on a predictive model, exhibits significantly superior overall performance compared with traditional cascaded controllers. This paper presents a systematic review of MPDSC methods for PMSMs. Within a unified predictive control framework, existing research is categorized into four types: finite control s‐et MPDSC (FCS‐MPDSC), deadbeat‐based MPDSC (DB‐MPDSC), continuous control set MPDSC (CCS‐MPDSC) and explicit MPDSC (EMP‐DSC). Focusing on key problems, representative research works and their improvement strategies are surveyed. After that, the two common issues of weighting factors and robustness are systematically discussed. Furthermore, a comparative analysis of typical methods is conducted to clarify their respective features and application scenarios. Finally, the main challenges and development trends of MPDSC in different terms are discussed. This review aims to provide systematic references and technical guidance for further research and practical applications of MPDSC methods.
Silicon carbide (SiC) power MOSFETs are gaining significant attention in various industries due to their superior performance in comparison to silicon-based MOSFETs. The reliability of discrete SiC power MOSFETs remains less explored compared to the SiC power modules. This paper investigates the reliability of the SiC power MOSFETs through DC power cycling, using TO-247 packaged devices as the representative for the discrete devices. The evolution of electrical (R ds-on, V th) and thermal (R th-jc) parameters for four Wolfspeed SiC MOSFETs (C3M0016120D) with gradual ageing is tracked till their end of life (EOL) is reached. After the EOL is reached, non-destructive X-ray computed tomography is used to correlate the physical changes within the package to the changes observed in the electrical and thermal properties. Experimental results showed that R ds-on, V th, R th-jc, and junction temperature can be used for the condition monitoring of SiC MOSFETs in reliability-critical applications like aerospace and electric vehicles, R ds-on demonstrating the highest sensitivity and practical utility as a condition-monitoring parameter under the specific test conditions, package structure, and dominant failure mechanisms investigated in this work. The degradation trends are driven by electro-thermal stress on aluminium bondwires and the die-attach solder layer.
Press-pack IGBTs are widely used in high-voltage and high-power applications due to advantages such as double-sided heat dissipation and ease of series connection. However, the parallel operation of multiple chips inside the device easily leads to uneven current distribution. Therefore, it is significant to investigate non-invasive current sensing methods for ensuring device reliability. This paper focuses on the non-invasive sensing of internal current distribution in press-pack IGBTs, analyses electromagnetic interference caused by busbars in converter-valve application scenarios, and proposes a corresponding correction method. First, the structure of the converter valve and the position of the side busbars around the device are analysed. Then, the electromagnetic interference of the side busbars on current sensing is investigated, and a correction method is proposed. Simulation results show that the calculated angle is 45.22 degrees in the absence of busbars and 44.32 degrees after applying the proposed correction method in the presence of busbar interference. Finally, a double-pulse experimental platform is built for validation. Under the upward-biased condition, the proposed method corrects the offset angle from 300.15 degrees to 350.25 degrees, and under the leftward-biased condition, it corrects the offset angle from 192.37 degrees to 258.25 degrees, thereby suppressing busbar interference and improving the accuracy of current-offset direction identification.
ABSTRACT The thyristor rectifier with two auxiliary converters of active power filters on the AC and DC sides can provide a large power supply for the DC load and a higher power quality. A circulating current exists between the main thyristor rectifier and auxiliary converters because the auxiliary converters have a common DC link. This article proposed a novel suppression method of the circulating current by adding an additional current loop in the DC side auxiliary converter. Initially, the reason for the circulating current is analysed, and then an additional current loop is added to the DC‐side auxiliary converter to eliminate the voltage that generates the circulating current. Compared with the traditional suppression method in which a zero‐sequence current control loop is used in the AC‐side auxiliary converter, the proposed control method has a higher utilisation of DC‐link voltage and a good compensation effect; thus, the proposed control method is more suitable for situations with limited utilisation of DC‐link voltage or lower DC‐link voltage of active power filters. To suppress the value of the circulating current to a smaller value, the Proportional‐integral (PI) plus Proportional Resonance (PR) regulator has been applied, and the suppression effect of the circulating current is improved.
ABSTRACT High‐gain DC‐DC converters play a vital role in photovoltaic conversion systems. Achieving high gain often requires operating at extremely high duty cycles, which can lead to increased costs, higher power losses, and reduced system performance. To address these challenges, this study proposes a non‐isolated semi‐quadratic DC‐DC converter, which is built based on the traditional CUK converter and operates in two distinct modes. In the first mode, the converter achieves a semi‐quadratic voltage gain ratio of D(3−2D)/(1−D) 2 , while in the second mode, the gain ratio is D/(1−D). A key feature of this design is the inclusion of inductive filters at both the input and output ports, ensuring continuous input and output currents. This reduces current stress on the output capacitors and minimises voltage ripple. The study evaluates the converter's performance in continuous conduction mode, analysing efficiency and small‐signal modelling while accounting for the effects of parasitic resistance under steady‐state conditions. Simulations were conducted using PLECS software, and the results are presented to validate the design. Additionally, a 75 W laboratory prototype was tested, with experimental efficiencies of 90.3% in boost mode and 12 W with experimental efficiencies of 88.1% in buck mode. These experimental results align closely with theoretical predictions, highlighting the converter's potential for practical applications. Due to its high efficiency, dual operating modes, and ability to deliver continuous currents, the proposed converter is well‐suited for renewable energy applications, particularly in PV systems and fuel cells.
ABSTRACT Phase‐shifted full‐bridge (PSFB) converters are widely used in medium‐ and high‐power applications due to their zero‐voltage‐switching (ZVS) capability. Conventional fixed dead‐time designs ensure ZVS under light‐load conditions but cause excessive dead time and increased body‐diode conduction losses at medium and heavy loads. This paper proposes a load‐current‐based closed‐loop dead‐time regulation method for PSFB converters, in which the dead time is dynamically adjusted in real time according to the measured output current. By correlating the required dead time with the resonant current level, the proposed approach forms a discrete‐time feedback regulation loop that updates the dead time on a cycle‐by‐cycle basis, ensuring sufficient ZVS margin while minimising unnecessary diode conduction. The control strategy is implemented using a digital signal processor without additional ZVS detection circuits, resulting in a hardware‐efficient solution with low implementation complexity. A 1 kW PSFB converter prototype is built to validate the proposed method. Experimental results demonstrate a significant reduction in excess dead time and body‐diode conduction losses, achieving an efficiency improvement of up to 0.2% compared with conventional fixed dead‐time control over a wide load range.
The modular multilevel converter with integrated supercapacitor-based energy storage system (MMC-SESS) has been considered promising to increase the stability of the renewable generation. However, power fluctuations in MMC-SESS operation decrease the power quality and stability of the grid connection. In this paper, a rolling-optimised fluctuation suppression strategy for MMC-SESS is proposed. The instantaneous arm power model is first established, incorporating dual-harmonic injection and associated parameter constraints. Based on this, the particle swarm optimisation algorithm is applied to obtain the best-so-far solutions of the main parameters in injected harmonics. Simulation verification shows that the proposed strategy can inject harmonics online as wind power fluctuates. It effectively suppresses arm power, submodule (SM) capacitor voltage, and SESS current fluctuations. It also extends the lifetime of the IGBT and supercapacitor. It also reduces the volume and cost of SM capacitors. These benefits demonstrate practical economic value for engineering applications.
To enhance the reliability of electrified railway traction systems (ERTSs), timely replacement of the DC-link capacitor is essential. Unfortunately, current DC-link capacitance estimation methods are often characterized by low accuracy and poor robustness to measurement noise. To address these issues, this paper presents an innovative DC-link capacitance estimation method that achieves high accuracy with a small experimental dataset. The variational autoencoder (VAE) algorithm, a generative machine learning technique, is used to effectively expand the dataset by generating new data, significantly reducing the cost and time associated with extensive experimentation. Additionally, the long short-term memory (LSTM) algorithm is employed to estimate the DC-link capacitance in ERTSs with high precision and strong resistance to measurement noise. The effectiveness of the proposed approach is validated through experimental results and analysis.To enhance the reliability of ERTSs, timely replacement of the DC-link capacitor is essential. This paper presents an innovative DC-link capacitance estimation method that achieves high accuracy with a small experimental dataset, which addresses the problems of the existing capacitance estimation approaches, i.e., low accuracy, poor robustness to noise and/or the requirement of a large amount of data.
Mitigating asymmetric aerodynamic loads while maintaining stable power output remains a significant challenge for large-scale wind turbines. This paper presents a parallel observation deviation-coupled linear active disturbance rejection control (PODC-LADRC) scheme for the pitch control of wind turbines. The pigeon-inspired optimization algorithm is incoporated to resolve parameter tuning complexities inherent in linear disturbance rejection control (LADRC). The proposed control strategy utilizes a dual-loop parallel observation architecture that overcomes the sequential tracking limitations of conventional LADRC, enabling near-zero delay disturbance estimation while maintaining better trajectory tracking accuracy. Performance evaluation via co-simulation with FAST and Simulink demonstrates substantial improvements across various wind regimes. Under turbulent wind conditions, the PODC-LADRC reduces average rotor moment and load variance by 58.4% and 38.2%, respectively, compared to conventional PI control. Furthermore, relative to conventional LADRC, the strategy achieves a 23.5% reduction in average moment and a 24.1% reduction in variance. Peak moment values are consistently attenuated across steady, step and gust conditions, thereby ensuring enhanced mechanical integrity and stable power generation.
