Addressing issues such as unbalanced DC-side capacitor voltage of the nine bridge arms and sub-module (SM) capacitor voltage fluctuations during the operation of the modular multilevel matrix converter (M3C), a circulating current control strategy for bridge arm voltage balancing is proposed. A mathematical decoupling model for both the input and output sides of the M3C is established. Through the analysis of the power of each bridge arm in the M3C, a hierarchical control strategy for bridge arm capacitor voltage balancing is designed, eliminating severe oscillations caused by low-frequency AC components of capacitor voltage ripple under the conditions f_m≈0 and f_m=f_g. A semi-physical experimental system for the M3C is built on the RT-LAB platform, and system operation under normal, low-frequency, and special operating conditions is simulated. Experimental results show that the proposed bridge arm voltage balancing control strategy performs well, with small fluctuation range of DC-side capacitor voltage and better output power quality.

The transient synchronous stability issue of voltage source converters is a typical nonlinear problem, which especially exhibits complex higher-order and interacting coupling characteristics in multi-converter systems. To clarify the interactive coupling characteristics and transient instability mechanism of the heterogeneous system with multiple control mode converters, a dual- converter hybrid grid-connected system consisting of grid-following voltage source converter （GFL-VSC） and grid-forming voltage source converter （GFM-VSC） is studied. Firstly, the mathematical modeling of the hybrid parallel system for transient stability analysis is developed using phasor analysis, and the mechanism of interactive coupling between two inverters is revealed. Then, by introducing three-dimensional surfaces and contour lines, a visual stability analysis method is proposed, which can intuitively describe the system balance relationship. Furthermore, the effects of the active power reference and reactive power control circuit of GFM-VSC and the current reference of GFL-VSC on transient stability are revealed in detail. The results show that the reactive power control introduction of GFM-VSC deteriorates the system balance relationship. The stability of the system can be improved by decreasing the active current injection of GFL-VSC and increasing its reactive current injection, or by decreasing the active power reference of GFM-VSC during grid faults. Finally, the hardware-in-the-loop experiments verify the correctness of the theoretical analysis.

In response to the issues of slow response speed and poor anti-disturbance performance of traditional control methods for three-phase Vienna rectifiers under system load variations and grid-side voltage fluctuations, as well as the problems of chattering and steady-state errors associated with commonly used sliding mode control, a control strategy based on second-order nonsingular terminal sliding mode and model predictive control is proposed. The second-order nonsingular terminal sliding mode control is introduced into the voltage outer loop to achieve rapid recovery and stable output of the DC-side voltage under load disturbances and grid voltage fluctuations. The model predictive current control is adopted in the inner loop to enable the input current to have fast tracking speed while maintaining low harmonic distortion. Simulation and experimental results verify the feasibility and effectiveness of the proposed control strategy.

Medium and high voltage direct current (DC) transformer is one of the key equipments for DC collection of new energy generation and voltage conversion in DC transmission and distribution system. Modular multilevel DC transformer (MMDCT) has high modularity and strong voltage expansion ability, which is the mainstream technical solution of high voltage and high power DC energy conversion, showing a rapid development trend. The typical application scenarios of medium and high voltage DC transformers are summarized, and then the typical topology of MMDCT is summarized. On this basis, the key technical problems of MMDCT such as modulation strategy, capacitor voltage balancing control method, circulating current suppression strategy, power control scheme, prototype and application, are analyzed and summarized. Finally, the development trends of MMDCT in the future are prospected, and the multi-functional roles and the multi-direction technical challenges of MMDCT in the future DC power grid are summarized.

In the field of direct current transmission, the new reinjection line commutated converter (RLCC) based on the theory of harmonic reinjection has significant advantages over the traditional line commutated converter (LCC), such as internal harmonic suppression ability, complete suppression of commutation failure, and dual-mode operation with leading and lagging trigger angles and so on. Aiming at the problem that the existing RLCC cannot decouple active power and reactive power with a single control variable firing angle, a new type of RLCC that can decouple power control is formed by adding freewheeling diodes to each reinjection branch of its shunt module and introducing modulation coefficient control variables. Based on topological structure and working principle, the theoretical current waveforms of AC and DC side are briefly analyzed. And the steady-state mathematical model of the new RLCC is derived. Based on the operating characteristics of the new RLCC, the working range of modulation coefficients, power operating range, and conditions for achieving unity power factor operation are determined. Finally, based on the current feedback control, a power decoupling control method is proposed combined with power factor feedforward control. The feasibility and correctness of the proposed control method are verified by simulation.

China’s power system is undergoing a profound transformation driven by the “dual-carbon” targets. With the large-scale integration of power electronics-based resources, system stability issues have become increasingly prominent, exhibiting strong coupling among multiple stability phenomena and unpredictable collective behaviors. To address these challenges, the limitations of the existing grid integration management framework is systematically analyzed, including the “black-box” external characteristics, inadequate grid integration testing system, and insufficient modeling accuracy. Drawing on international best practices, a new management framework is proposed, centering on the standardization of grid-connected control frameworks and model transparency. Comprehensive enhancement measures are proposed, covering the development of standardized control models and parameter tuning methods, technical standard revision, grid integration testing and simulation modeling optimization, digital monitoring, and market incentive mechanisms. The proposed framework aims to transform grid integration management from passive adaptation to proactive deep coordination, thus improving the secure and stable operation of new power systems.

Offshore wind power has the advantages of abundant resources and broad utilization prospects, and it is also of great significance for China to achieve “dual-carbon” goal. The tight space layout, harsh environment, and complicated operation and maintenance conditions of deep and distant sea offshore platforms require higher requirements for lightweighting, compactness, and reliability of voltage source converter based high voltage direct current (VSC-HVDC) converter stations. For the offshore wind power transmission system via VSC-HVDC, 6.5 kV power devices are used to reduce the construction cost and equipment footprint, and improve the transmission capacity per unit volume. A compact and lightweight VSC-HVDC power module is designed based on a 6.5 kV integrated gate commutated thyristor (IGCT), which improves the stability of the IGCT black start circuit and verifies its reliability through experiments, providing technical support for subsequent development.

A current loop improved model-free sliding mode control (IMFSMC) method is proposed to address the issues of steady-state power ripple and excessive current harmonics caused by dynamic loads, parameter mismatches, and single-phase short-circuit grounding faults in modular multilevel converters (MMC). Firstly, based on the mathematical model of MMC system, a model compensation hyperlocal model is constructed for parameter mismatch. Secondly, a time-varying gain adaptive sliding mode approach law is proposed to design a current loop sliding mode model-free feedback controller. Simultaneously, a model compensated improved linearly extended state observer (MCILESO) is designed to estimate unknown disturbances in the hyperlocal model and compensate them to the sliding mode feedback controller. The stability of the controller and observer is proved using Lyapunov function and Hurwitz theorem. Finally, based on MATLAB simulation and RTLAB semi physical platform, compared with PI control and traditional model-free sliding mode control, the results show that the proposed method can greatly improve the dynamic and steady-state performance of MMC systems under dynamic loads, parameter mismatches, and single-phase short-circuit grounding. Reduce the model dependency of MMC system, increase system robustness and anti-interference ability.

The mismatch risk and operational cost of off-line stability control strategy and setting values will increase dramatically with high proportion new energy generation which impose uncertainty on power system operation. Given that the fast adaptive online configuration of control strategy cannot replace the off-line strategy formulation procedure in practice for the moment, reforming the off-line stability control strategy configuration by a normative and automatic procedure is inevitable. A normative setting approach of power system off-line stability control strategy based on stochastic-robust mixed integer programming is proposed. The uncertain parameters are divided into two categories according to whether their probability distribution can be credibly provided, and incorporated by stochastic and robust programming respectively; the setting problem is normalized as a dual-branch bi-level stochastic-robust optimization. The proposed approach is demonstrated on a real stability control strategy of China Southern Power Grid, and its efficiency to support daily update is verified. This proposed approach provides a general normalization and automation technical framework for the security-sensitive and simulation-based off-line decision problems in an uncertain operational environment.

Aiming at the problem of arc suppression and protection of single-phase grounding fault in non effectively grounded distribution network, a single-phase grounding fault arc suppression method based on line voltage combination regulation of distribution network is proposed. Using the characteristics of single-phase grounding fault and line voltage constant in the regulation process, the three-phase line voltage is obtained from the adjustable injection transformer from the system, the injection transformer ratio and the secondary side switch are adjusted to flexibly regulate the zero sequence voltage, and the fault phase voltage is actively suppressed to 0, which realizes reliable arc suppression of grounding fault. Furthermore, a single-phase grounding fault type and feeder identification method based on the combined regulation of line voltage in distribution network is proposed. The change characteristics of injection current, zero sequence voltage and zero sequence current of each feeder before and after the change of injection transformation ratio are analyzed. The double identification criteria of ground fault type and feeder based on phase angle difference and admittance ratio are constructed to realize the sensitive identification of ground fault type and fault feeder. The proposed method is verified in PSCAD/EMTDC simulation environment. The simulation results show that the method can effectively suppress the grounding fault arc, realize the active, sensitive and accurate identification of grounding fault types and feeders, and improve the safe and reliable operation ability of distribution network.

The access of high proportion of new energy poses a challenge to the stability of the system, and it is of great significance to study the DC feeder system with asynchronous networking to ensure the large-capacity and stable transmission of the DC channel under the proportion of new energy. Introducing system voltage and frequency support capabilities, with the optimization goal of maximizing system external transmission capacity and new energy consumption capacity, and minimizing the imbalance of voltage support capacity, the DC external transmission capacity improvement optimization model based on the optimization of the sending-end system support capacity is established by introducing the system voltage and frequency support capacity. The model is divided into main and sub-problems, and the IMOPSO-PSD joint optimization simulation algorithm based on the improved multi-objective particle swarm algorithm is used to solve the model, and the optimal operation mode is screened out by the coefficient of variation method taking into account the imbalance of the voltage support capacity. Finally, the effectiveness of the proposed method is verified by taking the IEEE 39-node asynchronous DC sending-end grid model with high proportion new energy as an example.

To address the challenges of small targets, missed detections, and low image resolution in transmission line fault detection, a multi-scale feature fusion-based algorithm for transmission line fault detection is proposed. To obtain input features at different scales, images are first fed into a pre-trained shared residual network. Then, a feature fusion attention mechanism is used to learn salient features at different scales, integrating detailed information from large-scale feature maps and contextual information from small-scale feature maps. Additionally, inspired by dense spatial pyramid pooling in semantic segmentation, a multi-scale feature fusion dense pyramid is constructed to further enhance feature extraction capabilities. Finally, a scale-invariant error loss is utilized to predict depth mapping in logarithmic space. Experimental results on a transmission line fault detection dataset demonstrate that the proposed method achieves the highest detection accuracy, with a mean average precision (mAP) score of 97.50% and a detection speed of up to 56 frames per second. It exhibits high robustness and accuracy, providing an effective solution for intelligent monitoring of transmission line faults.

In the distribution network, the coordinated scheduling of flexible resources and distributed photovoltaics(PV) plays a crucial role in enhancing the consumption level of distributed renewable energy and promoting the realization of national "dual carbon" goals. However, traditional scheduling methods often require detailed topology and line impedance information of the distribution network, which is frequently unknown in actual distribution networks. To overcome this challenge, a non-parametric operation optimization method for distribution networks based on decision tree is proposed. This method first leverages unsupervised clustering to extract the topology status implied in the historical data to generate“pseudo labels”, and then train a decision tree to learn the relationship between operation strategies and the feasibility of flow constraints from historical operation data of the distribution network by training decision trees, and equivalently transforms the decision trees into mixed-integer linear forms that are easy to solve, thereby achieving a non-parametric reconstruction of flow constraints. Case studies show that this method can accurately depict the feasible domain of flow constraints under conditions where network topology and line impedance information are unknown, and its computational efficiency is significantly higher than that of traditional model-based methods.

Accurate ultra-short-term power prediction for distributed photovoltaic (PV) systems plays a crucial supporting role in providing essential information for distributed PV electricity sellers participating in the electricity spot market. Most existing methods employ cluster-based information-sharing modeling approaches to improve prediction accuracy. However, these methods either rely on direct data sharing, leading to serious data privacy issues, or depend solely on model interaction sharing.Although data privacy is ensured, the limited amount of shared information leads to poor prediction results. To address these challenges, a privacy-preserving, deep federated learning-based approach is proposed for ultra-short-term power prediction in distributed PV systems. Firstly, a federated learning framework based on dual-layer information interaction is constructed to enable richer information sharing through global model and global feature interactions among stations, avoiding direct data exchange to protect data privacy. Secondly, an attention-based autoencoder is adopted as the local prediction model to extract temporal features of power sequences, which are then coupled with global features to form spatiotemporal correlation features. By effectively mining and utilizing spatiotemporal correlation information, the power prediction accuracy is enhanced. Finally, simulation experiments using actual distributed PV data from a region in Hebei Province demonstrate the effectiveness of the proposed method.

Wind farms must be able to output active power according to grid dispatch instructions when connected to the grid. Due to the significant reduction in power electronic devices and the adoption of constant voltage-to-frequency ratio control operation principle, dynamic low-frequency wind power systems are more economical and reliable in the development prospects of deep-sea areas. To solve the problem that this system can only track the maximum power and cannot output active power according to grid dispatch instructions, a centralized variable-frequency wind farm active power control method considering active speed-up of the turbine cluster is proposed. By coordinating variable-frequency control and variable-pitch control, the problems encountered during control mode switching is solved. Comprehensively considering the operating and control characteristics of dynamic low-frequency wind power systems utilizing centralized transmission frequency, a suitable active power allocation method for wind farms is designed, which balances turbine rotatioanl speed and reduces pitch angle adjustments. Finally, the correctness and effectiveness of the proposed method are verified through PSCAD simulation models. The calculation results show that the proposed active power control method can effectively track grid dispatch instructions for active output of dynamic low-frequency wind farms, while also storing more kinetic energy, which helps to improve the control level of wind farms.

With the further development of energy storage and sharing economy, the operation mode of shared energy storage begins to be explored and paid attention to. In response to the problems of long investment and operation return cycles and low utilization efficiency in existing single energy storage scenarios, a shared energy storage investment and operation model is proposed mainly considering the three participants of shared energy storage operators, power grids, and users. Firstly, the operational models involving three parties are studied and analyzed, and a net profit model for shared energy storage operators, power grids, and users is established. Secondly, under certain profit conditions for the power grid and users, the goal of maximizing the revenue of shared energy storage operators is solved, and a mixed integer convex programming model is processed through quadratic fitting, M-envelope, and T-envelope methods. Then, to incentivize users to participate in energy storage, an improved Shapley value method is used to achieve partial grid delay benefit distribution among users. Finally, a simulation is conducted on three users in a certain region, and the Gurobi toolkit is applied to decide the investment and operation plan and service pricing of a shared energy storage operator, verifying the effectiveness of the investment and operation model.