Harmonic pollution sources in microgrids have the characteristics of high penetration and decentralization, as well as forming a full network. Local harmonic mitigation is a traditional harmonic mitigation method, which has the disadvantages of complexity and costly operation. Based on the idea of the decentralized autonomy of power quality, this paper establishes a comprehensive optimization model of the active power and harmonic mitigation capacities of grid-connected inverters based on two-layer optimization and realizes harmonic mitigation. Firstly, based on the harmonic sensitivity analysis, the calculation method of harmonic mitigation capacity is given. Secondly, a two-layer model of harmonic mitigation optimization is established. The upper-layer optimization model takes the minimum operation cost of the microgrid as the objective and the active power reduction in the multi-functional grid-connected inverter (MFGCI) as the optimization variable. The lower-layer optimization model offers the best harmonic mitigation effect as the objective and the harmonic current compensation as the optimization variable. According to the relationship between the total remaining capacity of MFGCI and the capacity required for harmonic mitigation, there are three different objective functions in the lower-layer optimization model. Then, the model solving steps are provided. Finally, an example shows that the proposed optimization model can achieve harmonic mitigation at different times. Compared with the case without active power optimization, the operation cost of the whole system can be reduced by up to 14.6%, while ensuring the harmonic mitigation effect. The proposed method has the advantages of a harmonic mitigation effect and economical system operation.

Fault diagnosis is one of the most important issues for a modular multilevel converter (MMC). However, conventional solutions are deficient in two aspects. Firstly, they lack the necessary feature information. Secondly, they are incapable of performing open-circuit fault diagnosis of the modular multilevel converter with the requisite degree of accuracy. To solve this problem, an intelligent diagnosis method is proposed to integrate the modal time–frequency diagram and FFT-CNN-BiGRU-Attention. By selecting the phase current and bridge arm voltage as the core fault parameters, the particle swarm algorithm is used to optimize the Variational Modal Decomposition parameters, and the fault signal is decomposed and reconstructed into sensitive feature components. The reconstructed signals are further transformed into modal time–frequency diagrams via continuous wavelet transform to fully retain the time–frequency domain features. In the model construction stage, the frequency–domain features are first extracted using the fast Fourier transform (FFT), and the local patterns are captured through a combination with a convolutional neural network; subsequently, the timing correlations are analyzed using bidirectional gated loop cells, and the Attention Mechanism is introduced to strengthen the key features. Simulations show that the proposed method achieves 98.63% accuracy in locating faulty insulated gate bipolar transistors (IGBTs) in the sub-module, with second-level real-time response capability. Compared with the recently published scheme, it maintains stable performance under complex working conditions such as noise interference and data imbalances, showing stronger robustness and practical value. This study provides a new idea for the intelligent operation and maintenance of power electronic devices, which can be extended to the fault diagnosis of other power equipment in the future.

The multi-port converter offers significant advantages in sustainable energy-based hydrogen production, such as high energy density and wide application potential. As the core energy hub in these systems, the multi-port hydrogen converter demands high reliability. However, its numerous switches and complex operation increase the risk of failure. Currently, there are no effective fault diagnosis or fault tolerance solutions for these systems. Current fault diagnosis methods for two-port and multi-port converters rely on many sensors, have low accuracy, and cannot handle multi-switch faults. Additionally, existing fault tolerance strategies fall short of maintaining stable power for the electrolyzer and don't allow capacity sharing among ports. This paper proposes a fault diagnosis method for open-circuit faults in a triple-port hydrogen converter and a corresponding fault-tolerant operating strategy. By optimizing a random forest algorithm with a sparrow search algorithm, high-precision multi-switch fault diagnosis is achieved using only three current sensors. Fault-tolerant operation of the system is ensured through an auxiliary switch following a failure. Additionally, power sharing between output ports can be realized. This ensures the efficient and stable operation of the hydrogen production system. The effectiveness of this solution has been validated through simulations and experiments.

Off-grid wind-to-hydrogen systems are considered a promising solution for sustainable, large-scale green hydrogen production in remote areas. However, under the combined effects of highly fluctuating wind generation and stochastic load variations, existing energy management methods still face a challenge: in off-grid wind-to-hydrogen systems, intelligent energy management studies that jointly address economic performance and operational stability are still limited. To address these issues, this paper develops a mathematical model for an off-grid wind-to-hydrogen system to reveal the coupling characteristics of the wind–electricity–hydrogen conversion process. Building on this model, a deep-reinforcement-learning-based energy management strategy is proposed. By formulating objectives that simultaneously capture economic benefits and stability requirements, the proposed strategy enables adaptive power flow allocation and dynamic optimization under uncertainty. Case studies demonstrate that, while fully satisfying load demand, the proposed strategy can significantly improve renewable energy utilization and hydrogen production, thereby increasing profit and ensuring stable and sustainable system operation.

This study introduces a data-driven modeling strategy for smart grid power quality (PQ) coupling assessment based on time series pattern matching to quantify the influence of single and integrated disturbance among nodes in different pollution patterns. Periodic and random PQ patterns are constructed by using multidimensional frequency-domain decomposition for all disturbances. A multidimensional piecewise linear representation based on local extreme points is proposed to extract the patterns features of single and integrated disturbance in consideration of disturbance variation trend and severity. A feature distance of pattern (FDP) is developed to implement pattern matching on univariate PQ time series (UPQTS) and multivariate PQ time series (MPQTS) to quantify the influence of single and integrated disturbance among nodes in the pollution patterns. Case studies on a 14-bus distribution system are performed and analyzed; the accuracy and applicability of the FDP in the smart grid PQ coupling assessment are verified by comparing with other time series pattern matching methods.

Current-source space vector amplitude modulation (SVPWAM) is an advanced and unique modulation technique that requires a well-suited regulation strategy for effective implementation. The traditional direct power control method for SVPWAM is quite complex, as it requires generating three-phase reference currents and a dc-link current to determine three different duty cycles. This results in the necessity for both the dc-link current control loop and the duty cycle selection process, thereby increasing the overall complexity. This letter proposes a simplified regulation structure that eliminates the duty cycle selection process and the need to sample the dc-link current, thereby removing the associated control loop while also ensuring adaptability to adverse grid conditions. Experimental results demonstrate that the proposed simplified approach achieves good performance in both the ideal and nonideal grid conditions.