The large-scale grid integration of distributed renewable energy enhances the flexible regulation capacity of the power system. However, the inherent randomness and volatility of its output, coupled with weak coupling access characteristics, pose severe challenges to the safe and stable operation of the power system. To address these issues, this paper proposes a power system planning method suitable for urban power grids. To accurately characterize the uncertainty of renewable energy output, the method incorporates the concept of multi-scenario stochastic optimization and introduces a dynamic scenario generation method for wind and solar power based on nonparametric kernel density estimation and standard multivariate normal distribution sequence sampling. This method generates a set of typical daily dynamic output scenarios for wind and solar power that closely match actual output characteristics. Considering the spatiotemporal response characteristics of flexible resources, the Soft Open Point (SOP) DC link enables flexible cross-node power transmission and spatiotemporal coupling regulation of flexible resources. Therefore, this paper constructs a mathematical model for the grid integration of flexible resources based on the SOP DC link. By integrating operational constraints such as power flow constraints in the power grid and source-load uncertainty constraints, a power system planning model is established. However, traditional convex optimization methods require approximate simplifications of the model, which can easily lead to a loss of accuracy. Although the Particle Swarm Optimization (PSO) algorithm is suitable for nonlinear optimization, it is prone to getting trapped in local optima. Therefore, this paper introduces an improved PSO algorithm based on refraction opposite learning, which enhances the algorithm's global optimization capability by expanding the particle search space and increasing population diversity. Finally, simulation verification is conducted based on an improved IEEE-39 bus test system, and the results show that the proposed scenario generation method achieves a sum of squared errors of only 4.82% and a silhouette coefficient of 0.94, significantly improving accuracy compared to traditional methods such as Monte Carlo sampling.

To address the issues of significant dc bus voltage and load fluctuations, as well as unstable power transmission in dual active bridge (DAB) converters within dc microgrid systems, this article proposes a segmented gain adjustment method based on multiplicative feedforward control (MFC-SGA). First, considering both steady-state and dynamic performance of DAB converters, two hybrid optimization control methods are proposed, and their advantages and disadvantages in terms of circuit parameter sensitivity and controller gain are analyzed. Second, to overcome the limitation of multiplicative feedforward control in light-load conditions due to restricted controller gain, the MFC-SGA method is introduced to enable adaptive parameter adjustment. Finally, an experimental prototype is built. Experimental results show that the MFC-SGA method is independent of inductance accuracy. When the operating condition changes, compared with the traditional method, the settling time is shortened by 60-83% and the overshoot is reduced by 37.5-62.5%; especially in light-load mode (10% of rated current), the dynamic response speed is improved by 68.75% compared with the MFC method, and the settling time is reduced from 32 ms to 10 ms. The experimental results verify the feasibility and effectiveness of the proposed method.

The spatiotemporal mismatch among computing demand, renewable energy supply, and electricity price levels poses significant challenges to the economic and low-carbon operation of geographically distributed data centers (DCs). While leveraging DCs' flexibility is promising, existing approaches often neglect the role of workloads' (WLs) logical constraints in migration scheduling and fail to fully exploit spatiotemporal variations in renewable energy availability and regional electricity prices to achieve simultaneous cost and carbon reductions. To bridge these gaps, this paper proposes a deep reinforcement learning (DRL)-based migration and scheduling strategy for multi-DCs participating in both electricity and carbon markets. Firstly, a unified dispatch framework enabling cross-DC WL migration is established to optimize the matching between WLs and renewable energy supply. Secondly, each WL is modeled by decomposing it into several dependent subtasks with logical constraints, enabling precise evaluation of spatiotemporal transfer potential. Finally, to address the computational complexity of large-scale, dependency-rich scheduling under dynamic market and renewable conditions, a proximal policy optimization (PPO) algorithm is proposed, leveraging its stability and efficiency for complex mixed-integer problems. Case studies demonstrate that the proposed strategy significantly reduces the DC operator's total operational costs by 4.1% compared to benchmark methods, effectively navigating dual-market dynamics and complex workload dependencies.

Accurate forecasting of distributed photovoltaic power plays a crucial role in ensuring the safety and stability of an active distribution network. However, most existing research on distributed photovoltaic power forecasting exhibits certain limitations, including: 1) insufficient consideration of the dynamic correlations among power sites; and 2) absence of a training loss function capable of simultaneously aligning the amplitude and shape of forecasting values with the true values. Therefore, a dynamic graph network with a shape-amplitude loss function based distributed photovoltaic ultra-short-term power forecasting method is introduced. Firstly, a data-driven method is used to mine the dynamic correlation and the dynamic graph data can be generated to ensure the effective characterization of the correlation among distributed photovoltaics. Secondly, the dynamic graph network is constructed as the power forecasting model to realize the effective utilization of spatial-temporal correlation features. Then, the shape-amplitude loss function which combines the Dynamic Time Warping and Mean Square Error is used as the criterion of model training to ensure the consistency of the forecasting value and the real in situ shape and amplitude. Meanwhile, the dynamic graph network is combined with quantile regression and the quantile loss function is improved inspired from the idea of shape-amplitude. The forecasting performance of the introduced approach is demonstrated via a distributed photovoltaic power dataset in China.

In recent years, the rapid integration of distributed photovoltaic (DPV) systems and electric vehicles (EVs) into distribution networks has posed significant challenges to grid security and stability due to their inherent output uncertainty and volatility. Grid operators can only access net load data from households by smart meters without direct recognition of installed distributed energy resources (DERs). Therefore, accurate recognition of user-side DERs is crucial for applications such as net load disaggregation, load forecasting, and demand response. However, existing recognition methods typically rely on extensive data for model training, exhibiting insufficient accuracy and robustness in few-shot scenarios. Furthermore, these methods often overlook the coupling effects between DPV and EV, which limits the recognition performance. To address these limitations, a two-stage recognition model of DPV and EV considering weather types and electricity consumption behavior is proposed. Firstly, weather-driven features are proposed for DPV recognition, and household travel pattern-based features are introduced for EV recognition. Secondly, an unsupervised DPV output power estimation algorithm is introduced to mitigate misidentification caused by the coupling of DPV and EV during EV recognition. Finally, a Light Gradient Boosting Machine (LightGBM) algorithm improved by multi-scale data reconstruction and consensus-weighted ensemble strategies is proposed to enhance model robustness and generalization capability under few-shot scenarios. Simulation results demonstrate that the proposed model achieves a recognition accuracy of 96% for both DPV and EV. This methodology provides an effective solution for the accurate recognition of DERs under conditions of resource coupling and limited data.

Green power chemical industrial parks represent a critical scenario for high-proportion renewable energy consumption and low-carbon transformation in high-emission industries. Addressing challenges in the park's multi-energy coupling system — encompassing wind/PV power, electrolytic hydrogen production, and chemical processes, such as random fluctuations at the source end, step disturbances at the load end, and strong cross-scale coupling between electricity, heat, and gas, this study conducts modeling and dynamic response quantification research. A nonlinear state-space model incorporating PEM electrolyzer polarization, thermal equilibrium, and multi-energy-flow interactions was constructed to analyze multi-timescale coupling mechanisms. A disturbance quantification method based on dynamic sensitivity and normalized variance ratio was proposed. Simulations demonstrate the model's capability to reproduce millisecond-scale electrical, second-scale transport, and minute-scale thermal evolution. The system exhibits a "low-pass filtering" effect due to the integrative effects of thermal inertia and gas storage. However, it exhibits inertial delay in responding to chemical load steps, with pressure recovery taking minutes. This research provides support for the safe, stable operation and flexible resource allocation of industrial parks.

The mismatch between the DC voltage control bandwidth and the low-pass filter control bandwidth results in a non-virtual inertia phenomenon in the DC voltage of the DC microgrid. For this purpose, a transfer function model of the DC microgrid is established in this paper, and the causes of the non-virtual inertia phenomenon are explained from the perspective of control bandwidth. Secondly, a virtual inertia response criterion based on control bandwidth matching is presented in this paper. Then, the concept and solution method of the control bandwidth matching domain are also provided in this paper. This control bandwidth matching domain can not only effectively ensure the virtual inertia characteristics of the DC microgrid but also be used to evaluate the system’s virtual inertia strength under different low-pass filter control bandwidths. Experimental results show that when the ratio of the voltage control bandwidth to the low-pass filter control bandwidth is greater than 10, the DC microgrid presents virtual inertia characteristics; otherwise, it exhibits non-virtual inertia (damped oscillation) characteristics.

To investigate the regulatory role of various types of energy storage systems in the distribution of active power within distribution networks, an optimal scheduling approach based on equivalent full-cycle lifetime modeling of lithium batteries is proposed in the paper. Electric vehicles are considered as representative flexible storage units, whose charging and discharging behaviors significantly influence the load profile of the network. By incorporating an equivalent full-cycle model, the method accurately quantifies the impact of varying depths of discharge on battery lifespan, and establishes a multi-objective optimization framework that accounts for battery degradation costs, load fluctuations, and peak-to-valley differences. A mixed-integer linear programming (MILP) approach is employed to simulate the scheduling process under a typical office building load scenario. The outcomes suggest that the proposed strategy effectively enhances the economic performance of energy storage systems, smooths the load curve, and optimizes the spatiotemporal distribution of active power. These findings offer a valuable analytical tool and optimization pathway for power regulation in distribution networks with high penetration of distributed photovoltaic systems.

With renewable integration and zero-carbon microgrids achieving 100% penetration, converter-dominated systems exhibit millisecond-timescale transient synchronization, which challenges existing physical cognitive methods and cognitive methodology with the synchronous generator (SG). In this paper, in order to quantificationally analyze the transient synchronization, a unified framework has been proposed that combines the generalized participation factor (GPF) method and basin of attraction (BOA) boundary analysis using the manifold approach. According to the GPF and BOA analyses, the fourth-order models are essential for accurate stability quantification, with synchronization controls (PLL, VSG, and droop control) contributing greater than 70% to transient dynamics versus about 20% from power-balance interactions. Further, the dynamic security region (DSR) is redefined by two typologies. Type 1 DSR maps stability in active-power injection space, and Type 2 DSR (generalized DSR) delineates limits in the controllable parameter space. The estimation procedures are proposed for these two types of DSRs by the BOA method. Finally, electromagnetic transient simulations and critical clearing time validation are employed for fidelity verification of models and estimation approaches. To sum up, the proposed novel framework enables systematic DSR estimations for renewable-rich power systems, empowering grid operators to optimize converter-controllable parameters and system operation conditions.

High penetration of distributed photovoltaic (PV) generation has transformed active distribution networks into inverter-dominated systems, where maintaining voltage stability, minimizing power losses, and maximizing renewable utilization under uncertainty remain significant challenges. Conventional centralized optimal power flow (OPF) and ADMM-based distributed optimization methods suffer from scalability limitations, high computational latency, and reliance on accurate system models, while single-agent reinforcement learning approaches such as PPO struggle with non-stationarity and lack of coordination in multi-inverter settings. To address these limitations, this paper proposes a coordinated control framework based on Multi-Agent Proximal Policy Optimization (MAPPO) for photovoltaic inverter clusters. By adopting centralized training with decentralized execution, the proposed approach enables effective coordination among heterogeneous inverter agents while preserving real-time autonomy. The framework explicitly incorporates network-level objectives, inverter operational constraints, and stochastic irradiance and load uncertainties, allowing agents to learn adaptive and robust control strategies. Simulation studies on a modified IEEE 33-bus active distribution network demonstrate that the proposed MAPPO-based method reduces voltage deviations by more than 40%, decreases network losses by approximately 25%, and lowers photovoltaic curtailment ratios by nearly 50% compared with centralized optimization approaches. In addition, MAPPO achieves significantly faster and more stable convergence than independent PPO under highly variable operating conditions.b These results indicate that MAPPO provides a scalable and resilient alternative to conventional optimization and single-agent learning methods, offering a practical pathway to enhance hosting capacity, operational robustness, and renewable integration in future active distribution networks.

High penetration of distributed renewables in the distribution grid has adverse effects on nodal voltage, network loss, and system stable operation. Essentially, the cause of these phenomena lies in the difficulty of the distribution network's reactive power to dynamically regulate and adapt to the varying power flow. Therefore, a multi-timescale reactive power optimization model for distribution networks considering network reconfiguration and the demand response (DR) of air conditioning (AC) systems is proposed in this paper. In the day-ahead reactive power optimization timescale, the On-Load Tap Changers (OLTCs) and capacitor banks (CBs) are modeled with a limited number of daily switching times, afterward are optimized with the network reconfiguration to alleviate power grid risk and minimize network loss, voltage deviation and operation cost of the distribution grid. Based on the day-ahead optimized operation statuses of OLTCs and CBs, in the intraday reactive power optimization timescale, the active and reactive power output of SVCs, SVGs, and photovoltaic (PV) inverters are flexibly controlled to achieve the minimum network loss and voltage deviations. To solve the nonlinear multi-timescale optimization model quickly and accurately, the adaptive immune particle swarm optimization (AIPSO) hybrid with embedded K-means strategy is designed to obtain the optimal tap positions of OLTCs and CBs as well as the optimized PV inverters' power output. Case studies based on a modified IEEE33-bus and standard IEEE118-bus distribution network validate the effectiveness of the proposed multi-timescale reactive power optimization strategy.

Accurate distributed photovoltaic power forecasting plays a crucial role in optimizing grid operations, enhancing economic benefits, and promoting the integration of new energy sources. However, existing methods for forecasting distributed photovoltaic power face several challenges: 1) Satellite cloud images can provide data support for distributed photovoltaics that lack specialized meteorological measurements, but the methods of cloud image features modeling tend to ignore important features; 2) Seasonal changes and variable climate conditions cause temporal distribution variations in photovoltaic output characteristics, leading to poor performance of trained forecasting models when there is a variation in data distribution, resulting in inadequate generalization capabilities. To address these issues, this paper proposes a regional ultra-short-term power forecasting method for distributed photovoltaic based on adaptive feature extraction and temporal transfer modeling. This approach integrates the spatial feature capture capability of Convolutional Neural Networks with the time-series processing mechanism of Transformer-based models to perform adaptive feature extraction of the correlation between multi-source remote sensing information and photovoltaic power. Subsequently, data distribution variations are quantified to divide the data into sequences with significant distribution differences. This allows the temporal transfer model to extract invariant generalized features in the temporal domain, thereby enhancing the model's generalization ability and forecasting performance. Finally, the effectiveness of the proposed method was validated using actual distributed photovoltaic power data.

Grid forimg (GFM) converters have been gradually promoted and applied in engineering practice owing to their ability to provide effective voltage and inertial support. With a high proportion of renewable energy sources being connected to the power grid, the grid-source interaction between the distributed power sources and power grid is further enhanced. To ensure the stable operation of the distributed power sources and suppress the instability of grid-source interaction in GFM inverters under a strong power grid, an improved line active disturbance rejection control (ILADRC) multiple parallel photovoltaic (PV) storage GFM virtual synchronous generator (VSG) system control strategy is proposed. First, impedance modelling of single and multiple parallel PV energy storage VSG systems is established for stability analysis using Gale circle stability criterion. Second, ILADRC method is designed, and impedance modelling of ILADRC multiple parallel PV energy storage GFM VSG system is established for stability analysis. Finally, an ILADRC multiple parallel photovoltaic energy storage GFM VSG simulation model and hardware-in-the-loop experimental platform are established for tests. Test results show that ILADRC multiple parallel photovoltaic energy storage GFM VSG system has better grid-source interaction stability under a strong power grid and grid harmonic background than parallel virtual impedance/unimproved method.

With the rapid development of renewable energy, more microgrids integrating photovoltaics and energy storage systems (MGPEs) have been deployed. Frequent grid faults have consequently increased the likelihood of MGPEs operating in off-grid mode, highlighting the need to maximize their autonomous operation. This paper proposes an optimal day-ahead scheduling method, which aims to maximize autonomous economic operation and minimize dependence on the main grid. Based on the autonomous results, the internal resources of MGPEs are aggregated into an equivalent power unit (EPU) and an aggregated energy storage system (AESS). The aggregated demands and residual capabilities of these aggregated models are then quantified, thereby facilitating coordinated scheduling by the main grid. Compared with traditional cost-optimization approaches, this case study shows that the proposed method reduces the daily outage probability of MGPEs by at least 24.9%.

With the increasing energy shortages and environmental degradation, the proportion of traditional fossil fuel-based power generation is gradually declining, being replaced by integrated energy systems that incorporate photovoltaic (PV) power generation, wind power generation, battery energy storage, and hydrogen-based power generation. However, PV power generation systems are highly influenced by environmental factors, exhibiting strong volatility and randomness, which creates challenges for grid dispatching. Conducting power forecasting for PV stations is thus essential to optimize grid operations. Yet, due to outdated communication systems in existing PV stations, integrated energy system cannot directly access operational data from all PV stations, making power forecasting difficult to implement. This paper proposes a method for predicting regional photovoltaic power generation using limited operational data from a few photovoltaic stations, based on long short-term memory (LSTM) network and graph neural networks (GNN). By deeply exploring the spatiotemporal characteristic relationships between these stations and the overall regional PV power generation, the method achieves data-driven, high-precision, and rapid power forecasting.

When grid causes transient fault, system performance will deteriorate. During LVRT period, grid-connected inverters will be affected by negative sequence components, second harmonic components, voltage drop, and over-current. To ensure system stable operation under power grid failure, this paper proposes based on virtual current +VPI PV GFL VSG LVRT control strategy. Virtual current control is introduced to eliminate over-current and current imbalance during faults, while introducing active power compensation to suppress active power reduction and meet load demand. To realize power stability control, a VPI resonant controller is introduced to suppress power doubling fluctuations. Parameter design rules and control performance of VPI resonant regulator are, respectively, analyzed. An experimental platform for PV GFL VSG power–current coordinated LVRT control strategy is established for experimental verification. Test results show GFL VSG power–current coordinated LVRT control strategy can not only suppress fault surge current and voltage drop during voltage sag faults, but also suppress the influence of negative sequence components and power doubling fluctuations, compared with conventional GFL VSG/GFL VSG positive and negative sequence separation LVRT control strategy, further improving system's LVRT capability. Finally, the feasibility of this method is further validated by applying it to photovoltaic energy storage micro-grid.

As power systems increasingly incorporate renewable energy sources and power electronic converters, the overall system inertia declines, making grid-forming energy storage systems essential for enhancing system stability and flexibility. This paper presents a modular electromechanical transient modelling approach for grid-forming energy storage systems. The proposed model comprises initial value calculation, power loop, injected current calculation, current limit compensation, and active/reactive auxiliary control modules, enabling precise simulation of the electromechanical transient behavior under various operating conditions and disturbances. Through simulations on a single-machine infinite bus system, the impact of frequency and voltage regulation parameters on both dynamic performance and steady-state levels is investigated. The results demonstrate that appropriate parameter selection can significantly improve the dynamic and steady-state responses of grid frequency and voltage under disturbances, providing valuable insights for the design and optimization of grid-forming energy storage systems.

In this paper, an IES optimal cooperative scheduling method based on a master–slave game is proposed considering a carbon emission trading (CET) and carbon capture system (CCS) combined operation with power to gas (P2G). We analysed the behaviour of integrated energy system operators (IESO) and energy suppliers (ES) when the system is operating in different states. This paper first introduces the structure of IES and the mathematical model of the game frame. Secondly, mixed integer linear programming and particle swarm optimization (MILP–PSO) are used. The final simulation results show that in the main scenario, IESO and ES have an income of CNY 181,900 and CNY 279,400, respectively, and the actual carbon emission is 106.75 tons. The overall income is balanced, and the carbon emission is in the middle. The results provide a reference value for operators and users to make decisions.

Existing methods for computing stability boundary of systems with multiple uncertain time delays are either conservative or suffer from low computational efficiency. Addressing this challenge, a value set based method is proposed to solve stability boundary for time delays with high efficiency. Firstly, the characteristic quasi-polynomial of the system is constructed in the periodic multiplication between frequency and time delays (FDM) space. Subsequently, the characteristic quasi-polynomial is transformed by the value set approximations, which is then refined through parameter space subdivision to achieve the required accuracy. Finally, the stability boundary is solved by traversing the parameter space with a bisecting step size. This method is characterized by its efficiency without introducing conservativeness, which is achieved through the decreasing step size space traversal. Numerical examples and a case study based on a microgrid with three distributed generators (DGs) that introduce communication delays are served to validate the accuracy and efficiency.Note to Practitioners-This paper proposes a value set based method to solve stability boundary of the system with multiple uncertain time delays. The method is developed based on the necessary and sufficient criteria. The method leverages the periodicity of FDM and employs a variable step traversal algorithm to decrease the computational burden and achieve high efficiency. Notably, this method can be used for systems with multiple time delays, regardless of the range or quantity of the delays. This method is effective for analyzing systems with multiple time delays, offering support for controller design and enhancing system stability.

To enable distributed PV to adapt to variations in power grid strength and achieve stable grid connection while enhancing operational flexibility, it is essential to configure grid-connected inverters with an integrated grid-following control mode, allowing smooth switching between GFL and GFM modes. First, impedance models of GFL and GFM PV energy storage VSG systems were established, and grid stability was analyzed. Second, an online impedance identification method based on voltage fluctuation data screening was proposed to enhance the accuracy of impedance identification. Additionally, a PV energy storage GFM/GFL VSG smooth switching method based on current inner loop compensation was introduced to achieve stable grid-connected operation of distributed photovoltaics under changes in strong and weak power grids. Finally, a grid stability analysis was conducted on the multi-machine parallel PV ESS, and a simulation model of a multi-machine parallel PV ESS based on current inner loop compensation was established for testing. Results showed that, compared to using a single GFM or single GFL control for the PV VSG system, the smooth switching method of multi-machine parallel PV ESS effectively suppresses system resonance under variations in power grid strength, enabling adaptive and stable grid-connected operations of distributed PV.