Based on a flexible DC distribution network with Full-Bridge Submodule Modular Multilevel Converters (FBSMMMC) and low-voltage-side Isolated Output Parallel Dual-Active-Bridge (ISOP-DAB) converters, traditional fault management relies on MMC blocking, causing complete network shutdown and limiting fast coordinated protection. Existing active fault identification and location methods also face constraints in current-limiting speed and fault information utilization. This paper proposes a hierarchical coordinated protection and diagnostic strategy enabling simultaneous response to DC line and low-voltage-side faults. In the early fault stage, it uses FBSM-MMC submodule voltage adjustment and closed-loop current control, combined with ISOP-DAB's bidirectional power flow, to achieve rapid staged current limiting. For post-limiting fault identification and distance measurement, the approach innovatively injects specific harmonic signals into the MMC under current-limiting conditions and applies Fourier analysis to extract characteristic components. By analyzing the returned signal's time delay and amplitude attenuation, the method calculates network admittance to quickly determine fault type and accurately locate the fault. This scheme supports multi-level current-limiting requirements while achieving accurate identification and location without MMC blocking, offering a new technical pathway to improve the safety and reliability of flexible DC distribution networks.

To address how variable user participation willingness affects electric vehicle （EV） dispatchable capacity assessment， this paper proposes a novel assessment method incorporating user willingness. First， joint entropy of probability density functions of EV load data is analyzed to identify dates with similar charging patterns， while a neural network predicts the probability distributions of initial charging times and initial state of charge （SOC） for future days. Second， a user participation willingness model is constructed by integrating four key factors—battery health status， remaining SOC， expected idle duration， and discharge electricity price incentives—using fuzzy logic rules to quantify response willingness. Finally， individual EV dispatch capacity is calculated subject to travel demand and idle duration constraints， followed by the development of an aggregate vehicle-to-grid （V2G） dispatchable capacity model for EV clusters. Case studies demonstrate that the proposed method accurately predicts EV charging demand， effectively quantifies user response willingness， and enables precise assessment of EV cluster dispatch capacity.

Accurate load forecasting is essential for ensuring the economic and reliable operation of integrated energy systems （IESs）. However， the nonstationarity， dynamically time-varying coupling， and high stochasticity of multivariate loads pose significant challenges to load forecasting. To address these issues， this paper proposes a probabilistic forecasting method based on multivariate variational mode decomposition （MVMD） and a dynamic hybrid graph attention network. First， MVMD is employed to synchronously decompose the original load sequences， enabling cross-series phase synchronization and modal alignment among multivariate loads. This process smooths the data while preserving the intrinsic temporal coupling characteristics between load series. Second， a dynamic hybrid graph attention network is designed. Through a gating mechanism， physical prior static graphs and a meteorology-driven dynamic graphs are adaptively fused， and spatiotemporal graph convolution is incorporated to extract higher-order time-varying coupling features. Finally， a quantile regression-based probabilistic output layer is constructed， extending conventional point forecasting to full probabilistic distribution forecasting. Experimental results based on the dataset from Arizona State University demonstrate that the proposed method effectively quantifies load uncertainty and significantly outperforms several benchmark models in forecasting performance.

To address the challenges posed by the widespread integration of distributed generation （DG） and electric vehicles （EVs） to the safe and stable operation of distribution networks， this paper proposes an optimal operation strategy for distribution networks that considers the support capability of DG and EVs. First， a multi-level control architecture based on a multi-agent system （MAS） is developed to enable information interaction and collaborative optimization among various agents. Second， by incorporating users’ travel demands， a credible capacity quantification model for charging and discharging stations is developed to assess their active support capability. Subsequently， a multi-objective optimal operation model is formulated， aiming to minimize operational cost， voltage deviation， and maximum line loading. This model is solved using a hybrid algorithm combining a genetic algorithm （GA） and an improved particle swarm optimization （IPSO）. Finally， operation simulations are conducted on a distribution system including renewable energy stations and various EV charging/discharging stations. The results indicate that the proposed method effectively reduces line loading， improves voltage profiles， and enhances the operational safety and economic performance of the distribution network.

Co-locating floating wind turbines (FWTs) with wave energy converters (WECs) offers a promising pathway to reduce the levelized cost of electricity (LCOE) for deep-sea renewable energy. However, stochastic platform displacements under coupled wind–wave–current loads aggravate wake interactions, degrade WEC shielding benefits, and increase cable fatigue risks—challenges not addressed by existing deterministic or empirical approaches. This paper proposes a bi-level distributionally robust optimization method for joint micro-siting and collector-system design of hybrid FWT–WEC plants. The upper level optimizes device layout by incorporating motion-aware wake–shielding coupling, while the lower level formulates a distributionally robust optimization (DRO) model for cable topology, sizing, and slack allocation using Wasserstein ambiguity sets constructed via kernel density estimation. Case studies demonstrate that the proposed method reduces wake losses by 53.70%, FWT heave amplitude by 42.3%, cable length by 4.81%, line losses by 12.76%, and reliability costs by 39.55%, achieving an overall LCOE reduction of 5.36% compared to conventional approaches.

Extreme ice storms cause severe icing faults on power lines and energy supply shortages, which can seriously threaten the safe operation of distribution systems. To enhance system resilience, a comprehensive resilience enhancement strategy for electric–hydrogen coupled distribution systems considering unmanned aerial vehicle (UAV) de-icing and hydrogen energy storage systems (HESS) is proposed in this paper. First, a line icing thickness model under extreme ice storms is established considering UAV de-icing effects to derive the corresponding fault rate. Second, resilience evaluation indicators are developed to quantify the disaster resistance and recovery capabilities of the electric–hydrogen coupled distribution system. Finally, a multi-dimensional disaster mitigation optimization model is formulated to strengthen the defense capability and resilience of the electric–hydrogen coupled system. Simulations are conducted on an improved IEEE 33-bus distribution system integrating UAVs and hydrogen energy storage systems, verifying that the proposed strategy significantly enhances restoration capability under extreme icing disasters while reducing economic costs.

Ice storms trigger transmission line icing and complicate emergency response， severely threatening the secure operation of integrated electricity-heat energy systems. To address this， this paper proposes a resilience enhancement method for integrated electricity-heat energy systems considering UAV de-icing without power interruption during disasters and post-disaster collaborative recovery. First， a line failure rate model accounting for ice-induced line losses quantifies disaster impacts. Second， resilience metrics are established across four dimensions： defensive capacity， adaptive response， coordination capability， and recovery effectiveness. Third， a multi-objective optimization model maximizes these resilience metrics， incorporating operational constraints of UAV de-icing， building thermal inertia， organic Rankine cycle （ORC） generation， and repair crew deployment. Finally， validation using a 6-node thermal system coupled with a modified IEEE 33-node distribution network confirms the method’s efficacy.

With the large-scale integration of renewable energy, the accuracy and dependability of the dynamic state estimation for modern distribution networks might be compromised by the uncertainty and fluctuation of renewable sources. To address these challenges, the paper proposed a new dynamic state estimation method based on Long Short-Term Memory instead of traditional Kalman Filter method. The proposed method exhibits promising accuracy with less time-cost by mitigating the negative influences of uncertainty and fluctuation brought by PV, all the while requiring limited measurements. In the paper, an improved photovoltaic power forecasting method was firstly introduced. The distribution network model considering PV forecasting effect was established by through the application of LSTM. Then, a dynamic state estimation method was developed based on established distribution network model. To prove the effectiveness of the method, the real time simulations based on RTDS platform and comparisons with traditional method were conducted.

To address the limitations of conventional voltage sag assessment metrics and the inability of existing methods to accurately characterize tolerance characteristic differences among various sensitive equipment types， this paper proposes a voltage sag severity assessment method that incorporates tolerance characteristics of sensitive equipment. First， the conventional severity metrics are improved by eliminating the upper constraints for voltage sag magnitude and duration in fault zones， thereby extending the assessment range. Second， magnitude-duration impact functions for equipment are constructed using the weighted function method， and a composite impact index is proposed. Building on this， an integrated evaluation model combining enhanced analytic hierarchy process （AHP） and coefficient of variation （CV） methods is developed to assess voltage sag severity for different sensitive equipment types at each node， balancing subjective and objective factors. Furthermore， the overall node voltage sag severity is evaluated by incorporating sensitive equipment proportions. Finally， simulations on the IEEE 30-bus system demonstrate that the proposed method effectively accounts for sensitive equipment tolerance differences and provides scientifically sound severity assessments for both equipment and nodes.

Modular Multilevel Matrix Converter (M3C) enables direct AC-AC conversion and has been widely studied in high-power applications such as railway traction, low-frequency power transmission. However, due to the lack of a DC link in the M3C, strong coupling between the input and output occur, leading to multiple control loops and a complex control system. Among the various control loops in M3C, the capacitor voltage balance control is a key component, as its performance directly affects output harmonic content and system stability. Traditional capacitor voltage balancing relies on feedback control, where the voltage of each capacitor in the submodule is sampled and fed back to the control loop to balance the capacitor voltages. However, as the number of submodules increases, the number of sampled capacitor voltages also increases, which complicates the capacitor voltage control loop and increases the cost. Therefore, this paper proposes a self-balancing method for the capacitor voltages in M3C. Without the need to sample the capacitor voltages of each submodule, capacitor voltage balance is achieved by selecting the appropriate switching matrix, thus reducing the control complexity and cost of M3C. Finally, a simulation and experimental platform for M3C is built to verify the feasibility and effectiveness of the proposed method.

With the rise of multi-energy prosumers (MEPs) in the local energy system, an efficient multi-energy management is increasingly significant. This paper designs a double auction electricity-heat market for simultaneous electricity and heat trading and proposes a bidding space model for MEPs. Firstly, it introduces the bidding rules, clearing mechanism, and market trading execution for the electricity-heat market. Secondly, a MEP’s energy supply cost model considering the equipment off-design performance and the market clearing price (MCP) prediction model considering the electricity-heat coupling are established, which together constitute the bidding space. Then, by analyzing the bidding space, it determines the optimal bidding quantity and a reasonable bidding price range, which guides MEPs to participate in the market. After that, considering the new bidding risks in the electricity-heat market compared to the traditional electricity market, it proposes an optimal bidding price model based on the conditional value at risk (CVaR), which helps find the most competitive price within the given range. Finally, case studies show that the proposed electricity-heat market and the bidding space model improve the economic benefits of the multi-energy system by 15.4%. Moreover, compared to the "cost-plus method" and "MCP prediction method", the proposed bidding strategy has an increase in revenue of 9.1% and 12.3%, respectively.

This paper presents an improved pitch control strategy for wind turbines (WTs) by integrating Model Predictive Control (MPC) with a Kalman filter to address control signal fluctuations under variable wind conditions. While MPC effectively handles system constraints and dynamic optimization, it often produces abrupt control outputs that can increase actuator wear and reduce system stability. To overcome this, a Kalman filter is applied at the output stage to smooth the control signal in real time. Simulation results on a three-turbine wind farm demonstrate that the proposed method not only improves pitch control smoothness but also enhances power generation and reduces mechanical loads. In a 300 -second test scenario, the strategy increases total power output by $8214.60 ~\mathrm{W} \cdot ~\mathrm{h}$ and significantly reduces tower base bending moment and spindle torque, confirming its effectiveness and practicality in wind turbine control applications.

Recurrence-based models inherently possess structural advantages in capturing temporal dependencies within wind power sequences, making them widely adopted in wind power forecasting applications. However, their strictly sequential computing mechanism leads to significantly constrained computational efficiency, thereby limiting their applicability in inertia support scenarios that demand both high precision and stringent timeliness. To overcome this limitation, we employ the Selective State Space Model (SSSM). This study is the first to apply this recurrent neural network that supports parallel recursion to the inertia support scenario, leveraging its structure for the rapid and precise modeling of temporal dependencies. Experiments on a real-world wind farm dataset show the proposed method achieves a significant $\text{5 8. 7 8 \%}$ improvement in timeliness while maintaining accuracy comparable to the state-of-the-art LSTM-based approach.

The dynamic changes of wind speed and direction have a significant impact on the power generation efficiency of offshore wind turbines (WTs). This paper proposes a coordinated yaw control of multiple offshore wind turbines especially considering the wind direction. First, on the foundation of Gaussian-Curl Hybrid model, we establish a wake model considering real-time dynamic wind direction. Then we propose Model Predictive Control based on this wake model to adjust the yaw angle of single WT, and establish a coordinated yaw control across multiple WTs on the basis of Serial Refinement which obtains WTs’ optimal yaw angle by ergodic search. Simulation results demonstrate that this approach effectively improves power generation while reducing fatigue loads.

DC transformers, as key equipment for collection and transmission, usually adopt an isolated topology structure. However, the intermediate frequency isolation transformer is constrained by magnetic materials and cannot achieve high power, making it difficult to meet the demand for large-scale offshore wind power transmission. Therefore, a non isolated DC transformer topology is proposed to cope with high-power scenarios. The T-type DC transformer is a new type of DC transformer structure with development potential in non isolated topology structures. This structure adopts a modular design approach, which is easy to expand and maintain. This article proposes a capacitor voltage self balancing strategy for a new type of T-type DC transformer. This strategy does not require a capacitor voltage sensor and is based on matrix theory. By presetting a switch matrix, the capacitor voltage is autonomously balanced, reducing the control complexity of the T-type DC transformer. Finally, the feasibility and effectiveness of the strategy were verified by establishing a simulation model of a T-shaped DC transformer.

It is particularly important to study the economic feasibility of energy transmission methods using hydrogen as a carrier and electric hydrogen hybrid transmission methods in response to the high construction, installation, and maintenance costs of electric energy transmission systems. Firstly, the advantages and disadvantages of electricity and hydrogen transmission modes are analyzed, the investment cost and benefit model of the single electricity and hydrogen transmission system is established, and the operation strategy of the single electricity and hydrogen transmission system is proposed; Secondly, for the multi-energy carrier delivery mode of electric-hydrogen hybrid, it is proposed to formulate the operation strategy of the hybrid delivery system by decoupling the system investment cost and operation income; Next, based on the cost model and operation strategy of each transmission system, plan and configure the transmission system to maximize its net profit. Finally, the actual data of a certain offshore wind power plant was used for calculation and analysis, and the economic performance of five transmission systems was compared and analyzed. The calculation results showed that the hybrid transmission system of flexible direct current transmission and LH2 ship transmission of hydrogen energy had the best economic efficiency.

The double auction market is an effective way to realize P2P energy trading in the local multi-energy system. However, in the double auction market, the electricity and heat trading are often auctioned separately, which brings the difficulty of energy decoupling to multi-energy prosumers (MEPs) and increases their trading risk. Therefore, this paper designs an electricity-heat coupling double auction market (EHC-DAM), where the electricity-heat combination is regarded as a single commodity. Firstly, it introduces the local multi-energy system framework and the operation mode of EHC-DAM. Then, it proposes an "electricity-heat-price" threedimensional supply (demand) surface for MEPs to participate in the market bidding. The bidding form of the 3D surface fully reflects the energy supply capacity and the expected economic benefit of MEP. After that, it establishes an electricity-heat coupling market clearing model aimed at maximizing social welfare, which determines the trading quantity and price of electricity and heat. And for its constrained discrete nonlinear characteristic, an improved Parallel Artificial Protozoa Optimizer (PAPO) is proposed to solve it. Finally, case studies demonstrate that the proposed EHC-DAM reduces the total energy cost of the local multi-energy system. Moreover, compared with PSO, GA and APO, the improved PAPO has stronger competitiveness.

With the implementation of the carbon peaking and carbon neutrality strategy, the proportion of new energy in the power system continues to increase, while the frequency regulation ability of the system gradually decreases. To address this problem, this paper studies the participation of distributed integrated energy microgrid group in frequency regulation auxiliary service. First, based on the concept of microgrid individuals participation in frequency regulation market after aggregation, the mode of integrated energy microgrid participating in frequency regulation auxiliary service market is described. Then, a cyber-physical-social system (CPSS) of integrated energy microgrid group is established. Based on this model, the frequency regulation cost of microgrid group is calculated. The cost model considers multiple frequency regulation modes, and classifies the social attributes of microgrid by using the grey weighted clustering method to simulate and analyze the willingness and probability value of microgrid individuals to participate in the frequency regulation market. Afterwards, based on Monte Carlo simulation, the cost calculation method for microgrid aggregators to participate in the frequency regulation auxiliary service is proposed, which provides the basis for aggregators to participate in the frequency regulation market quotation. Finally, an aggregator with five integrated energy microgrids is simulated and analyzed, and the curve of its frequency regulation cost is obtained. The simulation results show that the impact of the social attributes of the individual microgrid on the frequency regulation cost of the aggregator can reach 6.921%, so the uncertainty risk caused by the social attributes of the microgrid should be fully considered in the bidding process.

Integrated energy microgrids (IEM) have emerged as an effective way to improve energy efficiency and promote distributed energy utilization. IEM systems acquire electricity and gas from external markets and supply electricity/heat/cold to users. In this paper, we study the optimal energy purchase strategy for IEM, considering the impact of demand response incentives. Firstly, considering the uncertainties, we construct an IEM medium- and long-term market multi-energy purchase model based on conditional value-at-risk, optimizing the portfolio of electricity and gas purchases, as well as their proportion in total energy amount. Subsequently, based on medium- and long-term daily energy supply curves and day-ahead load forecast results, a spot market energy purchase model is established to optimize the spot purchase of electricity and gas, maintaining the supply-demand balance while minimizing operating costs. Furthermore, we design demand response incentives and develop a master-slave game model between IEM and users to guide the formulation of the energy purchase strategy by incorporating corrected load data as feedback. The energy purchase strategies are resolved by the GUROBI solver, while the optimization of demand response incentives is carried out through the PSO algorithm, all based on the MATLAB platform. The adaptability of the proposed model and strategy is verified.

The integration of stochastic renewable energy sources into energy systems presents challenges due to flexibility insufficiency and uncertainty modeling inaccuracy. To address these challenges, this paper explores a high-dimensional covariance matrix approach based on stochastic differential equations (SDEs) for energy and reserve co-dispatch in integrated electricity and heating systems. By capturing the long-term correlation of wind power forecast errors, the SDE model simultaneously models point forecasts and high-dimensional positive-definite covariance matrices, resulting in a compact uncertainty set. To ensure computational efficiency, we employ duality theory to transform the stochastic co-dispatch problem with the high-dimensional uncertainty set into deterministic convex programming. Additionally, a virtual heat storage model is introduced for the district heating network, leveraging the thermal inertia of the network to enhance reserve capacity. Simulations validate the calibration and sharpness of the proposed high-dimensional uncertainty set. The results demonstrate that co-dispatch using the SDE-based uncertainty set reduces operational costs by 4.9% compared to a set that does not consider the long-term correlation of wind power, under a 95% coverage rate. Moreover, the storage capabilities of district heating pipelines and smart buildings provide cost-free reserves, saving the operational costs of the two test systems by 2.8% and 1.3%, respectively.