IET Power Electronics最新文献

With the improvement of the new energy penetration rate, the power grid presents the characteristics of a weak current network, and a series of unstable problems appear. Compared with the grid-following inverter (GFLI), the grid-forming inverter (GFMI) provides essential voltage and frequency stabilisation capabilities for weak grids and has received extensive attention. Under the condition that the grid frequency does not exceed the limit, in order to give consideration to the economy, it is hoped that the inverter can track the maximum power point (MPP) of new energy so as to improve the utilisation rate of new energy. However, the active power control bandwidth of GFMI is low, which cannot meet the requirements of MPP tracking control rapidity. Therefore, this paper proposes a fast power control strategy for GFMI: firstly, the active power control bandwidth is improved by a prefilter with leading characteristics; then, model reference is used to adaptively adjust the parameters of the prefilter to avoid the impact of grid impedance changes on bandwidth enhancement performance. With the proposed control, the GFMI can quickly adjust the output power and track the maximum power point of new energy without affecting the stability. Finally, experimental validation confirms the technical feasibility and performance superiority of the proposed strategy.

This paper presents heuristic domain reduction finite control set model predictive control (FCS-MPC) for multilevel current source inverters that addresses the computational burden with improved performance. The proposed strategy is designed by taking advantage of the characteristics of the current source inverter and integrating its structural features and electrical properties to narrow the range of switching state options. By reducing the computational load from exponential to quadratic order, the proposed method significantly reduces the computational burden on the controller while maintaining system performance, making it more suitable for real-time applications. Simulations and hardware-in-the-loop testing demonstrate performance comparable to conventional FCSs across various operating conditions, while its reduced complexity enables higher sampling frequencies without CPU overload. Whereas the conventional approach is limited by CPU overload, the suggested approach allows for improved system performance at higher sampling rates or with more intricate control schemes.

To address the issues of limited noise resistance capability in fault diagnosis and inherent difficulties in fault isolation within conventional model-driven methods, an inverter open-circuit fault diagnosis method based on residual evaluation and machine learning is proposed in this paper. Firstly, the linear Kalman filter is designed to estimate three-phase currents subject to the noise interference. Current residuals are calculated by subtracting the estimated values from actual fault currents. Secondly, the residual evaluation function based on the p-norm is introduced. Subsequently, the residual evaluation features within each time subdomain are computed in sequence. Meanwhile, the alarm threshold via weight negative correlation is put forward. Fault diagnosis is achieved by comparing evaluation features with the threshold. Thirdly, the feedforward neural network is employed to extract nonlinear features of current residuals. Feature components after principal component analysis are input into the support vector machine to accomplish classification tasks and locate failure power transistors. Simulation results show that the fault diagnosis scheme performs the fast diagnosis speed and has the ability to avoid false alarms. The fault isolation scheme based on the machine learning framework achieves precise location of the failure power transistor on the basis of rapid diagnosis.

Current research on sensorless control for induction motors (IMs) with silicon inverters shows inadequate performance at low speeds. However, there is no reported literature on the influence of silicon carbide (SiC) device characteristics on the performance of IM sensorless control for model-based methods, and the influence remains unclear. This article addresses this gap by demonstrating an IM sensorless control system with SiC inverters and investigating its speed estimation accuracy under low-speed conditions. Firstly, the mechanism relationship between SiC device characteristics and speed estimation accuracy is revealed to reduce estimation error. Further, a three-level adaptive compensation method, considering non-ideal factors, such as the electromagnetic interference caused by high dv/dt of SiC devices and the parasitic capacitance effect, is proposed to improve the speed estimation accuracy. As a result, the low-speed operating range is extended to 15 rpm (0.5 Hz), and the speed error is below 5 rpm in dynamic and static conditions of the speed from 30 to 60 rpm, which is verified on a 7.5 kW IM drive platform based on the SiC inverter.

Traditional power generating is harmful to the environment; hence, renewable energy generation is becoming more popular. The performance of the converters that connect such power generation sources to the grid is critical. Multi-level inverters are used in solar-based photovoltaic applications as they offer better performance, structural flexibility and isolated inputs. To overcome the harmonic difficulties, the switching instances of these inverters are estimated offline utilising sophisticated computational approaches. Even though several optimisation tactics have been tried in the past, the switching frequency, modulation index, or both have been considered as fixed parameters. A 3Φ 11-level cascaded multi-level inverter was chosen for analysis in this study because it has been widely documented in studies on optimum switching instance determination. The best switching angles for various modulation indexes and switching frequencies were determined using five computational methodologies. The PV system with MLI has been developed in the MATLAB/Simulink environment and tested. The outcomes are compared with the existing work as well as the IEEE-519:2014 standard. The findings back up the inverter's better performance due to optimised switching employing the latest intelligent computational algorithms.

DC microgrids (DCMGs) play a pivotal role in integrating renewable energy sources into modern power systems. However, constant power loads (CPLs), such as motor drives and electronic devices, introduce negative incremental impedance (NII), posing serious stability challenges. This research addresses the destabilizing effects of CPLs in DC-DC boost converters (DBCs) by proposing a robust composite control strategy that enhances system stability and dynamic performance. The control framework integrates global integral terminal sliding mode control (GITSMC) with the backstepping method, while the secretary bird optimization algorithm (SBOA) is employed to eliminate manual tuning and ensure precise regulation of DC-bus voltage. A virtual capacitor is introduced to mitigate the low-inertia effect, and exact feedback linearization transforms the nonlinear DBC model into Brunovsky's canonical form, effectively addressing both NII and non-minimum phase behavior. An exponential reaching law further reduces chattering and ensures fast convergence, while Lyapunov theory confirms global stability. Simulation results demonstrate 52.15% faster settling time, 83.63% lower overshoot, and 86.87% reduced undershoot. Compared with BS-SMC, SOSMC, and conventional SMC, the proposed controller achieves up to 81.74% improvement in settling time and up to 95% reduction in overshoot and undershoot. Experimental results further validate the controller's superior transient response and robustness, making it well-suited for CPL-dominated DCMGs.

This paper proposes an LLC converter circuit with capacitor voltage balancing control for wide output voltage range applications. This proposed topology consists of a three-level half-bridge LLC structure on the primary side, which helps reduce the voltage stress on power devices and is suitable for high input voltage supply. It is operated in a complementary 50% duty cycle of fixed switching frequency for DC-DC transformer (DCX) mode and boost mode and operated in PWM for buck mode. The secondary side employs an active-bridge rectifier with phase-shift control to boost the output voltage, thereby extending the voltage regulation range. This design enables a wide range of voltage regulation capabilities and ensures a smooth transition between modes. In addition, considering two capacitors connected in series at the input of the three-level circuit that may cause voltage imbalances due to component mismatch or asymmetrical switching times of power devices, the study also proposes a control strategy with the designated control circuit to maintain the voltage balance of capacitors. To validate the effectiveness of this method, a circuit with a 400 V input, an output ranging from 150 to 400 V, and a power range of 1 kW has been implemented to realise the feasibility of the proposed approach.

The major challenges with solar photovoltaic-based grid-connected systems are poor sensitivity, poor selectivity, improper voltage control and poor power quality parameters like total harmonic distortion (THD), due to which synchronisation between SPV and the grid is difficult to maintain. In the proposed research, the solar photovoltaic is connected to the grid via a boost converter and inverter. The perturb and observe (P&O) method is used for extracting the maximum power from SPV and boost converter controlling. The existing controllers have complex mathematics and occupy a computational burden. In order to overcome such a problem, inverter controlling has been assessed with the implication of a novel design of mixed poison distribution and an artificial neural network (M-POS-ANN) controller. Such a novel design of controller is simple with less data computation and easy to apply in comparison to existing methods. The controlling of the inverter is also assessed with the cuckoo search algorithm (CSA) and fuzzy logic controller (FLC). The proposed M-POS-ANN controller improves the selectivity to 0.082 p.u., enhances the sensitivity to 5.3 p.u., provides a minimum total harmonic distortion of 3.4% and the least DC link voltage deviation of 2.9% in comparison to CSA, FLC and other existing methods. The effectiveness of the M-POS-ANN controller is also tested and validated in the hardware setup of a 2 kW solar module, a 3 kW boost converter, a 5 kW inverter on an STM32F407 controller.

Submodule (SM) failure is one of the key factors leading to instability in modular multilevel converter (MMC) systems. To address this challenge and enhance system reliability, redundant configurations have become the core solution. However, existing redundancy strategies cannot simultaneously balance the transient repair speed and steady-state output accuracy. Hence, a novel fusion redundant control scheme that combines cold-standby and hot-standby redundancy strategies is proposed to overcome the challenge of optimizing both transient repair speed and steady-state output accuracy. First, an MMC structure is proposed, which is equipped with both cold-standby SMs and hot-standby SMs and realizes the sharing of cold standby SMs. Then, the casting and switching strategies of the SMs are designed based on this structure. Finally, a bridge arm reference voltage update module is designed to avoid the voltage offset caused by the coordination of the cold-standby module and the hot standby module. Simulation and experimental results confirm that the proposed method effectively addresses both transient and steady-state performance issues, facilitates rapid repairs, and ensures precise and reliable operation of the MMC system after a fault event.

This paper introduces a high-gain voltage quadratic boost converter (QBC) and for control presents a non-linear control method based on introduction and damping assignment passivity based control (IDA-PBC) to regulate it. The proposed controller overcomes the challenges in this nonlinear converter and improves the converter's performance against variations in load, input and reference voltage. To eliminate the steady state error, an integrator is included in the IDA-PBC structure. Furthermore, the genetic algorithm is employed to determine the control coefficients for improved controller performance. To evaluate the effectiveness of the proposed controller, A simulation is conducted in the preliminary stage using Matlab/Simulink. The simulation results confirm its robust performance in the presence of disturbances such as input voltage and load changes and it is shown that the rise and fall times as well as the amount of overshoot are improved compared to the LQR and PI controller minimiser. Subsequently, the proposed control scheme is brought in to practice for QBC and experimental result indicate the applicability of this strategy.