The integration of distributed energy resources (DERs) has escalated the challenge of voltage magnitude regulation in distribution networks. Model-based approaches, which rely on complex sequential mathematical formulations, cannot meet the real-time demand. Deep reinforcement learning (DRL) offers an alternative by utilizing offline training with distribution network simulators and then executing online without computation. However, DRL algorithms fail to enforce voltage magnitude constraints during training and testing, potentially leading to serious operational violations. To tackle these challenges, we introduce a novel safe-guaranteed reinforcement learning algorithm, the DistFlow safe reinforcement learning (DF-SRL), designed specifically for real-time voltage magnitude regulation in distribution networks. The DF-SRL algorithm incorporates a DistFlow linearization to construct an expert-knowledge-based safety layer. Subsequently, the DF-SRL algorithm overlays this safety layer on top of the agent policy, recalibrating unsafe actions to safe domains through a quadratic programming formulation. Simulation results show the DF-SRL algorithm consistently ensures voltage magnitude constraints during training and real-time operation (test) phases, achieving faster convergence and higher performance, which differentiates it apart from (safe) DRL benchmark algorithms.

@en

The topology of low-voltage distribution networks (LVDNs) is crucial for system analysis, e.g., distributed energy resources (DERs) integration, network hosting capacity analysis, state estimation, and electric vehicle charging management. However, it is frequently unavailable or incomplete. This paper develops a data-driven topology identification approach for LVDNs with a high proportion of underground cables. The proposed approach exploits the fact that underground cables usually follow the street pattern, thus relying on open street map (OSM) and smart meter (SM) data. Three stages compose the proposed approach: In the first stage, a hierarchical minimum spanning tree algorithm is proposed to generate the initial topology with an accurate number of sub-branches from the pre-processed OSM data and peak demand. In the second stage, based on the limited SM data, the location of breakpoints in mesh topology caused by circle roads is verified and reconstructed to guarantee the radial structure of LVDNs. Finally, given multiple incomplete SM datasets, three data-driven optimization models based on a state estimation model are constructed to mitigate the error of cable length induced by OSM data. The feasibility of the proposed topology identification approach is verified on three actual LVDNs in The Netherlands and multiple incomplete SM datasets. Furthermore, the minimal amount of SM data needed to minimize the error of cable length is analyzed.

@en

This study presents an innovative approach to risk-aware decision-making in water resource management. We focus on a case study in the Netherlands, where risk awareness is key to water system design and policy-making. Recognizing the limitations of deterministic methods in the face of weather, energy system, and market uncertainties, we propose a scalable stochastic Model Predictive Control (MPC) framework that integrates probabilistic forecasting, scenario generation, and stochastic optimal control. We utilize Combined Quantile Regression Deep Neural Networks and Non-parametric Bayesian Networks to generate probabilistic scenarios that capture realistic temporal dependencies. The energy distance metric is applied to optimize scenario selection and generate scenario trees, ensuring computational feasibility without compromising decision quality. A key feature of our approach is the introduction of Exceedance Risk (ER) constraints, inspired by Conditional-Value-at-Risk (CVaR), to enable more nuanced and risk-aware decision-making while maintaining computational efficiency. In this work, we enable the Noordzeekanaal–Amsterdam-Rijnkanaal (NZK-ARK) system to participate in Demand Response (DR) services by dynamically scheduling pumps to align with low hourly electricity prices on the Day Ahead and Intraday markets. Through historical simulations using real water system and electricity price data, we demonstrate that incorporating uncertainty can significantly reduce operational costs—by up to 44 percentage points compared to a deterministic approach—while maintaining safe water levels. The modular nature of the framework also makes it adaptable to a wide range of applications, including hydropower and battery storage systems.

@en

Deep Reinforcement Learning (DRL) presents a promising avenue for optimizing Energy Storage Systems (ESSs) dispatch in distribution networks. This paper introduces RL-ADN, an innovative open-source library specifically designed for solving the optimal ESSs dispatch in active distribution networks. RL-ADN offers unparalleled flexibility in modeling distribution networks, and ESSs, accommodating a wide range of research goals. A standout feature of RL-ADN is its data augmentation module, based on Gaussian Mixture Model and Copula (GMC) functions, which elevates the performance ceiling of DRL agents, achieving an average performance improvement of 21.43%, 1.08%, 2.76%, by augmenting five-year, one-year and three-month data, respectively. Additionally, RL-ADN incorporates the Tensor Power Flow solver, significantly reducing the computational burden of power flow calculations during training without sacrificing accuracy, maintaining voltage magnitude with an average error not exceeding 0.0001%. The effectiveness of RL-ADN is demonstrated using distribution networks with size varying, showing marked performance improvements in the adaptability of DRL algorithms for ESS dispatch tasks. Furthermore, RL-ADN achieves a tenfold increase in computational efficiency during training, making it highly suitable for large-scale network applications. The library sets a new benchmark in DRL-based ESSs dispatch in distribution networks and it is poised to advance DRL applications in distribution network operations significantly. RL-ADN is available at: https://github.com/ShengrenHou/RL-ADN and https://github.com/distributionnetworksTUDelft/RL-ADN.

@en

This paper investigates the impact of adaptive activation functions on deep learning-based power flow analysis. Specifically, it compares four adaptive activation functions with state-of-the-art activation functions, i.e., ReLU, LeakyReLU, Sigmoid, and Tanh, in terms of loss function error, convergence speed, and learning process stability, using a real-world distribution network dataset. Results indicate that the proposed adaptive activation functions improve learning capability over state-of-the-art activation functions. Notably, adaptive ReLU activation shows the most improved learning process, with convergence speed up to twice as fast as ReLU. Adaptive Sigmoid and Tanh activation functions exhibit superior performance in terms of loss function error, outperforming ReLU and LeakyReLU by up to two orders of magnitude. Furthermore, the proposed adaptive activation functions lead to smoother and more stable learning processes, especially during early training, improving convergence. The practical implications of this study include the potential application of these adaptive activation functions in distribution network modeling, forecasting, and control, leading to more accurate and reliable power system operation.

@en

Electrical power systems are witnessing a paradigm shift from traditional synchronous generators towards an in-creased integration of power electronic interfaced (PEl) generation. As the global community leans towards renewables, ensuring grid stability during this transformation becomes paramount. This paper presents a fundamental study of oscillatory stability dynamics for three emerging grid-forming converter controller topologies: Virtual Synchronous Machine (VSM), droop control, and the Synchroverter, in comparison to conventional syn-chronous generation. Utilizing the two area 4 generator (2A4G) system for analysis, the research underscores the significance of converter integration, proximity-based enhancements in damping capabilities, and the delicate equilibrium in parameter tuning for optimal stability. The results pave the way for informed decision-making in grid development and renewable energy com-prehension, highlighting the potential of grid-forming converter controllers in steering a sustainable energy future.

@en

Anticipating failures is vital for maintaining a reliable power supply. Advanced measurement devices in the grid generate vast data that contains valuable information on grid operations. Initial signatures of an incipient failure are often reflected in this data in the form of electrical waveform distortions. Conventional protection schemes are not equipped to analyze these distortions and anticipate failures. There is a considerable research gap for a simple yet robust and universal failure anticipation and diagnosis scheme. This paper proposes a universal Failure Anticipation and Diagnosis Scheme (FADS) to detect incipient failures in AC distribution grids. The method comprises three short stages, helping the operator make an informed decision. In the first stage, the FADS scheme leverages the fundamental properties of electrical sinusoid waveforms to detect distortions. In the second stage, the distortion data is processed through pre-determined thresholds set in accordance with the system's regular operation. In the third stage, depending on the system, the FADS uses the extent of the violations of these thresholds and ranks the severity of the danger posed to grid operations. The classification helps determine if the waveform distortions are the signature of an incipient failure. The proposed FADS method's reliability, robustness and effectiveness are evaluated in incipient failure conditions of field events modelled in real-time simulations on standardized IEEE distribution feeders. The FADS is a high-speed distortion detector, is quite sensitive, and the method has high selectivity because of its nature.@en

This paper explores the potential application of quantum and hybrid quantum–classical neural networks in power flow analysis. Experiments are conducted using two datasets based on 4-bus and 33-bus test systems. A systematic performance comparison is also conducted among quantum, hybrid quantum–classical, and classical neural networks. The comparison is based on (i) generalization ability, (ii) robustness, (iii) training dataset size needed, (iv) training error, and (v) training process stability. The results show that the developed hybrid quantum–classical neural network outperforms both quantum and classical neural networks, and hence can improve deep learning-based power flow analysis in the noisy-intermediate-scale quantum (NISQ) and fault-tolerant quantum (FTQ) era.@en

It is challenging to determine the active filter control factors in wind power plants (WPP) to obtain effective harmonic voltage mitigation and avoid over-modulation or system instability problems caused by overlarge feedforward or feedback gains. To address this issue, an active filter tuning (AFT) method is proposed in this paper to offer a common parameter tuning and stability assessment strategy for both current-controlled and voltage-controlled harmonic impedance reshaping (HIR) methods by introducing a coordination factor including different weight coefficients for system stability margin, wind turbine (WT) harmonic suppression, and grid harmonic mitigation. The superiority of the proposed method is verified in a 20 MW WPP with four cases considering both WT harmonic voltage amplification and grid voltage amplification. Compared to previous methods, the AFT-based HIR method achieves a more flexible balance between system stability and harmonic voltage mitigation, obtaining better performance in WT harmonic suppression, grid harmonic suppression, system robustness, and transient response.

@en

The increasing proportion of renewable energy introduces both long-term and short-term uncertainty to power systems, which restricts the implementation of energy management systems (EMSs) with high dependency on accurate prediction techniques. A hierarchical online EMS (HEMS) is proposed in this paper to economically operate the Hybrid hydrogen–electricity Storage System (HSS) in a residential microgrid (RMG). The HEMS dispatches an electrolyzer-fuel cell-based hydrogen energy storage (ES) unit for seasonal energy shifting and an on-site battery stack for daily energy allocation against the uncertainty from the renewable energy source (RES) and demand side. The online decision-making of the proposed HEMS is realized through two parallel fuzzy logic (FL)-based controllers which are decoupled by different operating frequencies. An original local energy estimation model (LEEM) is specifically designed for the decision process of FL controllers to comprehensively evaluate the system status and quantify the electricity price expectation for the HEMS. The proposed HEMS is independent of RES prediction or load forecasting, and gives the optimal operation for HSS in separated resolutions: the hydrogen ES unit is dispatched hourly and the battery is operated every minute. The performance of the proposed method is verified by numerical experiments fed by real-world datasets. The superiority of the HEMS in expense-saving manner is validated through comparison with PSO-based day-ahead optimization methods, fuzzy logic EMS, and rule-based online EMS.

@en

One crucial aspect of Modular Multi-level Converter (MMC)- Bipolar Point-to-Point (BPP) configuration systems is the occurrence and damping of oscillations on the DC side of HVDC networks. These oscillations can arise due to various factors, including the interaction between the AC and DC systems, de-blocking of converters after a fault, and the dynamic behaviour of connected power sources. Various investigations consider damping methods that delve to mitigate oscillatory tendencies and establish stability during Post-Fault (PF) recovery. However, the current research on damping predominantly focus on the impact of AC fault or unbalanced conditions on the DC side. This paper presents an investigation that addresses the gap concerning with sub-synchronous oscillations occurring during the de-blocking of a MMC-BPP within the post-DC fault recovery. The investigation also considers active damping and enhanced current control loop as a combined mitigation measure. A meticulous real-time digital simulation supported parametric sensitivity analysis is conducted on a four-terminal MMC-BPP synthetic power system. Numerical results provide insight into the level of effectiveness that can be achieved by the considered concept of active damping.

@en

Probabilistic modelling of power systems operation and planning processes depends on data-driven methods, which require sufficiently large datasets. When historical data lacks this, it is desired to model the underlying data generation mechanism as a probability distribution to assess the data quality and generate more data, if needed. Kernel density estimation (KDE) based models are popular choices for this task, but they fail to adapt to data regions with varying densities. In this paper, an adaptive KDE model is employed to circumvent this, where each kernel in the model has an individual bandwidth. The leave-one-out maximum log-likelihood (LOO-MLL) criterion is proposed to prevent the singular solutions that the regular MLL criterion gives rise to, and it is proven that LOO-MLL prevents these. Relying on this guaranteed robustness, the model is extended by adjustable weights for the kernels. In addition, a modified expectation–maximization algorithm is employed to accelerate the optimization speed reliably. The performance of the proposed method and models are exhibited on two power systems datasets using different statistical tests and by comparison with Gaussian mixture models. Results show that the proposed models have promising performance, in addition to their singularity prevention guarantees.

@en

Power grid digitalization introduces new vulnerabilities and cyber security threats. The impact of cyber attacks on power system stability is a topic of growing concern, which is yet to be comprehensively analyzed. Traditional power system stability analysis is based on the impact of non-malicious small, and large physical disturbances. However, cyber attacks introduce a new dimension to power system stability, in which malicious cyber actors can selectively target critical systems and applications and cause severe stability issues. Hence, in this work, the traditional disturbances considered in power system stability classification are expanded from physical to cyber-physical disturbances caused by cyber attacks. Based on a thorough state-of-the-art, an analysis of how cyber attacks can translate into physical disturbances affecting the traditional power system stability categories is performed. The system stability analysis is expanded by mapping the power system stability categories with the defined cyber-physical attack types. The findings of this work showcase the importance of cyber security for power system stability.@en

To enhance the economic viability of integrated energy systems, it is important to balance risk and reward while ensuring operational flexibility and compliance with regulatory constraints. Developing an integrated risk measurement method for energy trading in energy markets that complies with regulatory constraints adds another layer of complexity, especially when dealing with large-scale heterogeneous energy systems, such as those dominated by green hydrogen. Ensuring adherence to market regulations, grid codes, and flexible requirements requires a thorough understanding of legal frameworks and policy considerations, as well as integrating regulatory constraints into the bidding/offering strategies. To address these challenges and enable the profitable development of integrated electricity-hydrogen energy systems (IEHSs), this paper develops a risk-constrained bidding/offering self-scheduling strategy for IEHSs that allows for simultaneous trading in day-ahead energy and reserve electricity markets. The developed self-scheduling strategy regulates hydrogen and power flow in a centralized manner, taking into account the operational limits of the power sector, hydrogen fueling stations (HFSs), electric vehicle parking lots, fuel cell vehicle parking lots, and renewable energy sources. As a salient feature, the important human-related factors in the mobility sector are included in the proposed strategy to motivate private owners to collaborate with IEHSs and to accommodate the charging demand of vehicles with electricity market signals. To obtain the robust day-ahead schedule, economic risks associated with renewable power generation, calls for reserve deployment, and electricity market prices are taken into consideration. Towards this end, a stochastic adaptive robust optimization approach is proposed to control the risk of profit variability, which is solved by a nested column-and-constraint generation algorithm. The effectiveness and feasibility of the proposed strategy are verified using the modified IEEE 14-bus test system, as a standard scale of local IEHSs. The numerical results highlight the synergies impact between HFSs and the power sector, resulting in an increase in the profit of the IEHS from $101.654k to $275.078k. Furthermore, the IEHS can improve its profit margin and competitiveness by employing the tri-level stochastic adaptive robust optimization approach. This approach ensures that the IEHS is resistant to sudden changes in real-time interactions, even with an 8.14% increase in the imbalance cost.@en

Power systems are undergoing rapid digitalization. This introduces new vulnerabilities and cyber threats in future Cyber-Physical Power Systems (CPPS). Some of the most notable incidents include the cyber attacks on the power grid in Ukraine in 2015, 2016, and 2022, which employed Advanced Persistent Threat (APT) strategies that took several months to reach their objectives and caused power outages. This highlights the urgent need for an in-depth analysis of APTs on CPPS. However, existing frameworks for analyzing cyber attacks, i.e., MITRE ATT&CK ICS and Cyber Kill Chain, have limitations in comprehensively analyzing APTs in CPPS environments. To address this gap, we propose a novel Advanced Cyber-Physical Power System (ACPPS) kill chain framework. The ACPPS kill chain identifies the APT characteristics that are unique to power systems. It defines and examines the cyber-physical APT stages spanning from the initial phases of infiltration to cascading failures and a power system blackout. The proposed ACPPS kill chain is validated with real-world APT attacks on the power grid in Ukraine in 2015 and 2016, and cyber-physical simulations.

@en

Fault ride-through capability studies of MMC-HVDC connected wind power plants have focused primarily on the DC link and onshore AC grid faults. Offshore AC faults, mainly asymmetrical faults have not gained much attention in the literature despite being included in the future development at national levels in the ENTSO-E HVDC code. The proposed work gives an event-triggered control to stabilize the system once the offshore AC fault has occurred, identified, and isolated. Different types of control actions such as proportional-integral (PI) controller and super-twisted sliding mode control (STSMC) are used to smoothly transition the post-fault system to a new steady state operating point by suppressing the negative sequence control. Initially, the effect of a negative sequence current control scheme on the transient behavior of the power system with a PI controller is discussed in this paper. Further, a non-linear control strategy (STSMC) is proposed which gives quicker convergence of the system post-fault in comparison to PI control action. These post-fault control operations are only triggered in the presence of a fault in the system, i.e., they are event-triggered. The validity of the proposed strategy is demonstrated by simulation on a ±525 kV, three-terminal meshed MMC-HVDC system model in Real Time Digital Simulator (RTDS).

@en

Anomaly detection is of considerable significance in engineering applications, such as the monitoring and control of large-scale energy systems. This article investigates the ability to accurately detect and localize the source of anomalies, using an autoencoder neural network-based detector. Correlations between residuals are identified as a source of misclassifications, and whitening transformations that decorrelate input features and/or residuals are analyzed as a potential solution. For two use cases, regarding spatially distributed wind power generation and temporal profiles of electricity consumption, the performance of various data processing combinations was quantified. Whitening of the input data was found to be most beneficial for accurate detection, with a slight benefit for the combined whitening of inputs and residuals. For localization of anomalies, whitening of residuals was preferred, and the best performance was obtained using standardization of the input data and whitening of the residuals using the zero-phase component analysis (ZCA) or zero-phase component analysis-correlation (ZCA-cor) whitening matrix with a small additional offset.

@en

Sensor attacks on grid-tie photovoltaic (PV) inverters can cause severe damage. Considering uncertain environments and unknown model mismatches, real-time estimation and defense for sensor attacks on actual PV inverters are challenging. In this article, we propose an optimization-driven robust estimator within the attack frequency range using the H_∞ index, while the model mismatch effect on estimation is also minimized. To improve the real-time response under varying environments, an analytical solution from a convex quadratic programming reformulation is constructed. Guided by the estimation, we further develop a closed-loop compensation strategy with a tracking controller and a low-pass filter. Through code porting, our proposed defense strategy has been implemented in a microcommercial PV inverter. Hardware implementations show that our defense approach can effectively mitigate sensor attacks and maintain stable inverter operation.

@en

Cascading failures in power systems are extremely rare occurrences caused by a combination of multiple, low probability events. The looming threat of cyberattacks on power grids, however, may result in unprecedented large-scale cascading failures, leading to a blackout. Therefore, new analysis methods are needed to study such cyber induced phenomena. In this article, we propose a data-driven method for dynamical analysis of power system cascading failures caused by cyberattacks. We provide experimental proof on how attacks may accelerate the cascading failure mechanism, in comparison to historically observed blackouts. Using a dynamic power grid model, consisting of multiple, coordinated protection schemes, we define and analyze the point of no return in a cascading failure sequence by applying the Hilbert-Huang transform for time-frequency analysis. Numerical results indicate, cyberattacks may accelerate cascading failures at least by a factor of 3x. This is due to the excitation and non-damping of multiple frequency modes greater than 1 Hz in a short time span. The proposed method is tested using time domain simulations conducted through a modified IEEE 39-bus test system, which can simulate cascading outages using coordinated protection schemes.

@en

Cyber security risks are emerging in Cyber-Physical power Systems (CPS) due to the increasing integration of cyber and physical infrastructures. Critical component identification is a crucial task for the mitigation and prevention of catastrophic blackouts. In this paper, we propose a novel method using graph data mining for critical CPS components identification named GraphCCI. First, it defines two categories of component correlations to reveal the cascading features of CPS. GraphCCI maps cascading failure datasets under time-varying operational states into weighted cascading graphs and constructs a graph database for graph data mining. By adopting graph data mining techniques, frequent subgraphs are identified to construct the Cascading Characteristics Graph (CC-Graph). Finally, the Node Criticality Index (NC-Index) is proposed to quantify the criticality of each CPS component. The experimental results on the IEEE 39-bus system verify the effectiveness of the proposed method and present an in-depth analysis of the CPS cascading features.@en