Heterogeneous distribution energy resources, such as renewable energy sources and electric vehicles (EVs) impose prominent challenges to active distribution network planning. This paper proposes a bi-level stochastic optimization framework that enables the coordinated optimization of renewable generation, flexible loads, grid capacity, and energy storage through an EV integration mechanism with rejection sets. A mixed-integer non-linear programming model is constructed to minimize total planning costs under technical and operational constraints, where upper-level investment decisions are coupled with lower-level operations. For the long-short term uncertainties in EV demand and renewable energy output, a scenario-based decision-dependent distributionally robust joint chance-constrained approach is employed for effective characterization. Finally, the proposed bi-level model is transformed into a single-level optimization problem via linearization, significantly reducing computational complexity. Simulation results on the 25-node system, IEEE 141 and IEEE 300 system demonstrate that the proposed approach significantly reduces the total planning cost and alleviates voltage violations and line congestion, while maintaining computational scalability.
Power systems are required to operate consistently within a safe frequency range. In this paper, we propose a safe optimal frequency control framework based on reinforcement learning to generalize and improve the classical linear droop-based primary frequency control. We formulate an optimal frequency control problem with frequency limit constraints and reformulate it with chance constraints. To solve this problem, we develop a reinforcement learning framework based on the twin delayed deep deterministic (TD3) policy gradient algorithm. This framework learns actor-critic neural networks for control policy design via primal-dual updates on a Lagrangian-form actor loss function, where the dual variables adapt to constraint violations while stabilizing the policy optimization. Simulations of a 39-bus test system verify that in response to disturbances in uncontrolled (e.g., renewable) generation or load, our framework effectively learns nonlinear decentralized controllers, which can stabilize the system faster and reduce oscillation compared to representative existing methods, while restricting frequency within limits.
Natural and human-induced disasters can trigger widespread outages in distribution systems, which makes resilience improvement a pressing concern. However, existing methods are inadequate due to unaddressed disaster uncertainty and overlooked short-circuit current limitations that underpin system stability. This paper develops a short-circuit current constrained preventive-restorative resilient control for distribution systems under disaster uncertainty, which is modelled as two-stage stochastic programming problem with conditional value-at-risk. The model jointly optimizes the preventive deployment carried out before a disaster and the restorative scheduling performed afterwards for mobile energy storage systems and other resources. The first stage fixes the optimal pre-disaster siting, while the second stage responds to faulted-branch topology changes and coordinates the charging and discharging profiles. This model considers short-circuit current constraints across both prevention and restoration stages. Case studies on a modified IEEE 33-node system reveal significant impacts of these short-circuit current limits. Validation on the IEEE 123-node system demonstrates the scalability of the proposed framework.
To address the difficulty of source-load balance matching in power systems with high penetration of renewable energy, this paper proposes a multi-time scale automatic "scenario-strategy" matching method for power system dispatch. Firstly, a scenario matrix containing the time series of system states and temporal constraints of flexible loads is constructed, and typical scenarios are identified and prioritized using multidimensional criteria. Secondly, a multi-objective dispatch optimization model is designed to generate flexible load strategy sequences adapted to state evolution. Finally, a nonlinear mapping between scenarios and dispatch strategies is established based on the LSTM model and deep learning. Case studies show that, compared with traditional single-time scale strategies, the proposed method can effectively improve dispatch performance and provide important support for secure operation and dispatch of power systems under massive and complex operation scenarios caused by uncertainties of renewable energy.
The increasing penetration of distributed renewable energy and the mass integration of electric vehicles (EVs) in distribution networks (DNs), particularly the proliferation of high-power EV superchargers, introduces sudden and high-power spikes that jeopardize scheduling efficiency and voltage stability. Consequently, there is an urgent need for rapid and accurate optimal scheduling methods to ensure safe DN operation. To address this challenge, a physics-informed voltage-aware deep reinforcement learning (PIVA-DRL) scheduling framework for DN optimal scheduling is developed. Firstly, to provide precise and robust real-time voltage perception faced with incomplete DN power flow model, a physics-informed compensated residual network (Pi-CORNet) that combines physics model with residual learning is proposed. Next, to achieve faster policy convergence and improved adherence to operational constraints, the Pi-CORNet is used to lead DRL agents to make informed decisions. Case studies conducted on the refined IEEE 141-bus DN demonstrate the proposed method can capture accurately voltage profiles under dynamic conditions and achieve faster convergence speed, thus improve the stability of DN operation.
Bidirectional asymmetric power transmission demand widely exists in DC microgrid with high distributed power penetration and custom-designing the bidirectional power rating of the DC converter rather than designing it symmetrically is beneficial to reduce the cost. However, soft-switching operation in the whole load range is challenging to existing schemes. In this paper, a near-zero switching loss asymmetrical isolated bidirectional DC converter (AIBDC) with two independent paths at the low voltage (LV) side is proposed, four silicon diodes are adopted to form forward path, while four fractional silicon carbide metal oxide semiconductor field effect transistors are used in the reverse path. With the proposed triangular modulation strategy, not only turn-on loss but also turn-off loss of switches can be eliminated significantly in the whole load range and near-zero switching loss can be obtained. Soft-switching characteristics of the proposed AIBDC with the considerations of dead time and voltage fluctuation are analyzed in detail. Parameters designing procedure and control strategy are given. To verify the effectiveness of the proposed AIBDC, a 280∼330V/120V/40kHz/2kW experimental prototype is built. The highest efficiency of the proposed AIBDC is 98.7%.
The renewable energy hydrogen system (REHS), which integrates renewable energy sources with hydrogen electrolyzer (HE) loads, is a key option for global decarbonization. However, as a converter-dominated system, the REHS is prone to small-signal instability under weak grid conditions and exhibits high sensitivity to operating-point variations. This paper analyzes the small-signal stability of the REHS and proposes an evaluation method to identify the stable operation region (SOR) under various operating conditions. First, an impedance model of the HE unit is derived considering the DC-side dynamics, and an impedance network model of the REHS is established to characterize source-load impedance coupling. The small-signal stability of the grid-connected REHS is further analyzed to determine how the operating conditions of HE loads influence system stability. Then, the SOR of the REHS is constructed by employing an efficient radial basis function-based fitting method to accurately determine the SOR boundary. Finally, case studies are presented to validate and further analyze the proposed method. The results show that the proposed method can clearly depict how the SOR boundary of the REHS is influenced by the grid strength, grid impedance angle, and generation-to-load capacity ratio.
With the increasing penetration of renewable energy, synchronous generators (SGs) and voltage source converters (VSCs) synchronized via phase-locked loops (PLLs) are expected to coexist in modern power systems, leading to complex dynamic interactions. Traditional small-signal stability analyses of PLL-VSCs often simplify or neglect the electromagnetic transients of SG stator flux and AC network dynamics, leaving the impact mechanisms of these dynamics on PLL-dominated stability modes largely unexplored. To bridge this gap, this paper establishes a comprehensive small-signal model for a parallel SG-VSC system. Within a self-stabilization and en-stabilization analytical framework, the complex torque paths associated with AC network dynamics and SG stator flux are rigorously extracted via branch decomposition. By analytically evaluating the complex torque coefficients under diverse operating conditions, the underlying mechanisms by which these dynamics modulate the PLL synchronization stability are unveiled. The findings reveal critical influence patterns involving the VSC capacity ratio, active power setpoints, PLL control parameters, and electrical distance. Specifically, it is revealed how the coupling between SG flux/ network dynamics and PLL alters the damping and synchronizing torque of the PLL-VSC.
Accurate and efficient spatiotemporal modeling is crucial for operating modern large-scale power systems, yet existing methods face limitations in scalability, task generalization, and computational complexity. This paper proposes a physics-guided block-sparse Spatiotemporal Graph Transformer that treats each node-time pair as a token and leverages electrical partitioning and temporal windowing to constrain complexity of attention mechanisms. Cross-block communication is restored via lightweight routing tokens over a physics-informed supergraph. A unified pretraining framework with masked token modeling and physics-aware objectives enables transferable representations for multiple downstream tasks. Evaluations on integrated transmission-distribution systems with up to 16,257 buses show the model outperforms representative spatiotemporal learning baselines in prediction accuracy, feasibility rate, and robustness to topology shifts and missing data, while reducing per-batch inference latency by 4–7x and peak allocated GPU memory by 60–70% versus standard Transformers. Code and data are publicly available at https://github.com/zyh1996saa/block_sparse_attn.
Integrating energy storage into modular multilevel converter (MMC) presents a promising solution for suppressing power fluctuations and riding through AC/DC faults. However, research on its application for frequency support in weak receiving-end grids remains limited. Additionally, existing frequency support strategies cannot achieve continuous and smooth coordination between virtual inertia control (VIC) and virtual droop control (VDC), which may lead to larger frequency deviations and slower frequency recovery. To address these issues, this paper proposes a novel frequency support strategy using supercapacitor-embedded modular multilevel converter (SC-MMC). The proposed strategy introduces weighting factors to adaptively adjust VIC and VDC contributions based on grid frequency. It imposes no additional computational burden on the control system. Simultaneously, it constructs smooth functions to modulate supercapacitor power output according to frequency deviation and SOC safety margin, thereby accelerating frequency recovery while avoiding output power jumps. Finally, PSCAD/EMTDC simulations are used to compare the proposed strategy with existing strategies under power deficits, and also to compare it with a 100% generator system under continuous load disturbances. The results demonstrate that existing strategies have various deficiencies, while the proposed strategy achieves the best overall performance, validating its effectiveness and superiority.
The increased frequency of extreme weather events and their growing societal impact necessitate the incorporation of humanitarian considerations into the restoration of distribution systems (DSs). However, existing research overrelies on static priority rules and cost-oriented strategies, neglecting both the inherent uncertainties of disaster evolution and the prioritized service of critical loads, resulting in delayed restoration of humanitarian services. This paper proposes a humanitarian-aware restoration framework for DSs that incorporates extreme weather-induced uncertainties and coordinates the dispatch of mobile emergency resources (MERs), including repair crews and mobile energy generators, etc. For the first time in DSs restoration, a shortage severity measure is introduced for bus-level evaluation of restoration adequacy, and a group-wise lexicographic optimization strategy is developed to sequentially minimize worst-case power shortages across prioritized humanitarian groups, thereby ensuring the timely restoration of critical loads, such as hospitals and emergency shelters. Then, a tractable co-optimization model of DSs and transportation networks is formulated with auxiliary variables to support MER routing decisions. Furthermore, a two-stage stochastic programming model is constructed to address uncertainties in line damage and repair times, and a progressive hedging-based distributed algorithm is employed to alleviate scenario-coupling computational burdens. Finally, case studies based on real data from the 2024 super typhoon “Yagi” demonstrate the effectiveness and superiority of the proposed approach.
This paper develops a coupled impedance model (CIM) and a coupling reshaping controller (CRC) for analyzing and mitigating sub-synchronous control interaction (SSCI) in hybrid renewable energy transmission systems. Unlike traditional aggregated impedance models, the proposed CIM captures not only the external interaction between aggregated subsystems and the power grid, but also the inner coupling among aggregated subsystems. This enables a comprehensive assessment of three aspects: the intrinsic damping provided by subsystems, the interaction damping induced by inner subsystem interaction, and the stability robustness issues caused by subsystem interaction. In addition, by reshaping the inner coupling among aggregated subsystems, the proposed CRC enhances the intrinsic damping while maintaining zero interaction damping, thereby effectively mitigating SSCI in hybrid renewable energy transmission systems. Furthermore, based on the stability robustness analysis results, the safe parameter region for the CRC is derived to ensure robust SSCI mitigation under model deviations. Impedance-based analysis shows that the proposed CRC increases the stability margin by 200% while keeping the sensitivity of the critical parameter below 1. Controller hardware-in-the-loop tests further verify the correctness of the CIM-based analysis method and the effectiveness and robustness of the proposed CRC.
With the rapid growth of renewable energy, power systems dominated by heterogeneous grid-forming (GFM) and grid-following (GFL) inverters are increasingly replacing conventional synchronous-generator grids, posing new challenges for stability assessment. The coexistence of diverse GFM and GFL inverters significantly complicates stability analysis, as inverter parameters deeply influence the system's frequency dynamics, yet traditional methods struggle to accurately evaluate these parameter effects. To address this issue, this paper proposes a dynamic model-embedded neural network approach to comprehensively assess the impact of inverter parameters on system stability. Firstly, a dynamic model that comprehensively accounts for mutual coupling effects among inverters is developed to reflect their impact on system dynamics. Secondly, this model is embedded into a task-oriented neural-network framework, enabling rapid, physically consistent, and interpretable stability assessment across varying parameter combinations. Finally, comprehensive analyses on the parameters of GFM and GFL inverters are conducted, highlighting the critical role of GFL inverter output capacity proportion in maintaining system stability, thereby providing valuable guidance for parameter optimization in practical renewable energy applications.
During metallic phase-to-ground faults, modular multi-level converter-based high-voltage direct current (MMC-HVDC) systems need to limit fault current to protect semiconductor devices. This paper proposes a novel bang-bang funnel control (BBFC)-based fault current limiting (FCL) strategy for MMC-HVDC systems. When overcurrent occurs in the AC side of an MMC-HVDC system, the controller switching mechanism is triggered, switching the regulator of grid-forming controlled MMC from vector control (VC) to BBFC. Then BBFC enables output current of MMC to rapidly track the predefined current reference, limiting the magnitude of fault current. Additionally, a feedback linearization control based on cascaded high-gain observer (CHGO) is employed to mitigate fluctuations in DC voltage when BBFC is activated. The BBFC-based FCL strategy demonstrates robustness against the variations of system parameters, active power, fault resistance and fault location. This method exhibits superior dynamic response performance under both symmetrical and asymmetrical faults. We validated the proposed FCL strategy using MATLAB/Simulink simulations and RT-LAB hardware-in-the-loop (HIL) experiments. The results demonstrate that the BBFC-based FCL strategy outperforms traditional fault current limitation methods in current limiting effect and robustness.
To overcome the limitations of existing pilot protection schemes, including slow operating speed and dependence on threshold tuning, this paper proposes a fast setting-less pilot protection scheme for multi-terminal direct current (MTDC) transmission lines in offshore wind power systems. First, a fault equivalent model for offshore wind power MTDC systems with a symmetric monopolar configuration is established. Based on this model, analytical expressions of the transient current difference (TCD) under internal and external fault conditions are derived. The analysis shows that TCD remains positive for internal faults and negative for external faults. Based on these opposite polarity characteristics, a fast setting-less pilot protection criterion with a mathematically determined threshold is developed. Extensive PSCAD/EMTDC simulations demonstrate that the proposed scheme is robust under challenging conditions, including high-resistance faults, noise interference, and communication disturbances. Moreover, it exhibits good adaptability across different DC line types and system topologies.
High penetration of inverter-based resources (IBRs) reduces system inertia and creates tighter interactions between frequency stability and transient stability. This paper proposes a robust stability-constrained optimization framework for IBR-dominated power systems. Instead of treating frequency and transient stability control as independent problems, the proposed framework coordinates them through shared dispatch decisions and IBR control parameter settings. The model jointly optimizes active power dispatch and control parameter settings of generators and controllable grid-following and grid-forming IBRs under renewable uncertainty. To solve this problem, a dual-loop solution framework is developed, where the inner loop solves the resulting two-stage robust UC problem using a Column-and-Constraint Generation algorithm, and the outer loop validates candidate dispatches and iteratively generates additional stability constraints through convex hull approximation and trajectory sensitivity analysis. The effectiveness of the proposed method is verified on modified New England 39-bus system and IEEE 118-bus system.
In the energy trading process, users form alliances to collectively trade energy with integrated energy service provider (IESP). However, the strategies of both parties influence each other, which may lead to revenue conflicts, while privacy-preserving concerns may hinder effective collaboration within the user alliance. To address these challenges, this paper proposes a game-theoretic energy trading strategy, in which alliance members exchange real-time information with IESP using standardized non-sensitive information. Specifically, an energy supply model for the hydrogen-based combined heat and power (CHP) system managed by IESP is established, and a demand response mechanism considering user alliance satisfaction is proposed. A master-slave game model is introduced to characterize the trading process, and a modified adaptive differential evolution with an optional external archive (JADE) algorithm is used to solve the energy trading and dispatching problem. Simulation results show that, compared with the interaction without considering games, the proposed strategy improves the revenue of the IESP by 27.0% and that of the user alliance by 1.0%, while also showing potential in carbon emission reduction and improved execution efficiency.
False data injection attacks (FDIA) targeting power systems will effectively evade bad data detection mechanisms and falsify relevant measurement data, thereby creating a grave menace to the operational stability and security of the entire grid. However, existing detection methods suffer from low convergence speed and inability when processing high-dimensional data, failing to accurately locate the specific attacked buses. To address these issues, we propose a novel FDIA location and detection approach for the FDIA, integrating principal component analysis with a radial basis neural network. A chaotic artificial immune algorithm is applied to optimize the radial basis neural network to classify the measurement data where the cubic chaotic mapping is incorporated into artificial immune algorithm to initialize the population so as to achieve faster convergence speed of the detection model. Meanwhile, the principal component analysis, which is computationally simple and retains the maximum information in the original samples, is utilized to rapidly downscale the high-dimensional historical measurement data, thereby enhancing the detection precision of the proposed method. To verify the effectiveness of the proposed method, we conduct simulations on the IEEE 14-bus and 57-bus systems. Comparative analyses with other detection methods are further performed to prove higher classification performance, higher accuracy, precision, recall and F1 value.
The large-scale integration of electric vehicles (EVs) into the power grid provides new flexibility resources for renewable energy accommodation. However, the randomness of their charging behavior and the uncertainty of renewable energy output pose challenges for aggregators participating in the electricity market. This paper proposes a two-stage scheduling of EV aggregators in electricity markets aimed at promoting renewable energy accommodation. First, a generalized energy storage aggregation model for EV clusters is constructed based on Monte Carlo sampling and Minkowski addition to accurately characterize the schedulable potential. Second, a bi-level optimization model for the day-ahead market is designed, considering the responsibility weights for renewable energy accommodation. Finally, in the real-time market stage, a rolling optimization model with the objective of minimizing future comprehensive operational costs is designed to dynamically correct the day-ahead plan and cope with the uncertainties of renewable energy and user behavior. Simulation results show that the proposed strategy can effectively maintain the grid's peak-to-valley difference ratio at an optimal 38.8%, avoiding the severe demand peaks caused by single-mechanism designs. Furthermore, it achieves a 39% relative improvement in the renewable energy accommodation rate compared to conventional price-guided orderly charging, achieving a synergistic optimization of aggregator economic benefits and clean energy accommodation goals while ensuring the safe and stable operation of the power grid.
The increasing penetration of electric vehicles (EV) in the Road Transport Network (RTN) and Distribution Power Network (DPN) coupling system tightly integrates the planning problems of both power and transportation systems. The interaction between power flow and traffic flow assignment presents a practical and significant challenge for the long-term planning of this coupled infrastructure. This paper investigates three key issues within the RTN-DPN coupled system: traffic flow assignment, fast charging station (FCS) planning, and energy storage (ES) system planning and scheduling for enhancing the load accommodation capacity of the coupled system. The proposed methodology addresses the optimal traffic flow allocation in the RTN, the planning of FCSs, and the optimal expansion of ESS in the DPN to host the charging demand from the RTN. A tri-level ES-FCS co-planning framework is modeled to support the RTN charging network and the DPN. First, accounting for the impact of road congestion on user routing decisions, a planning model for FCSs is established that integrates traffic flow, charger deployment, and charging load. This model aims to optimize the location and capacity of FCSs with the objectives of mitigating congestion in the RTN and reducing user travel time and cost. Subsequently, for the ES planning problem under multiple DPN performance indicator s, a Similarity to an Ideal Solution-Multi Objective Water Optimization Algorithm (SIS-MOWFO) is developed to coordinate ES planning in multi-objective scenarios. The proposed joint planning method is validated on a coupled network based on the IEEE-33 bus system and a coupled network based on the IEEE-118 system. Simulations demonstrate that: 1) Within the tri-level planning framework, the SIS-MOWFO algorithm improves convergence speed by 9.1%–19.4% and enhances the comprehensive evaluation score considering multiple indicators by 1.87%. 2) The optimal traffic flow allocation method, which considers congestion's influence on user routing decisions, reduces the road congestion indicator by 7.4% and decreases the average user travel distance by 3.26%. This work can provide a reference for multi-objective optimization in complex coupled energy system dispatch studies.
