Pub Date : 2025-07-02DOI: 10.1109/OAJPE.2025.3585425
Rohan Kamat Tarcar;Marjan Popov;Aleksandra Lekić
The increased use of High-Voltage Direct Current transmission networks requires appropriate control strategies for the converter stations, which are crucial to ensure uninterrupted energy supply. In this paper, a Polytopic Lyapunov Function-based Hybrid switching control strategy is implemented to combine the merits of Grid Following and Grid Forming control strategies by switching alternatively from one to another at the polytopes’ hyperplanes to ensure good system response even for faulty conditions. The state space equations of the control strategies are used to form the state hyperplanes for the switching rule. Since the hybrid switching control is based on the Polytopic Lyapunov Function, the system is inherently Large Signal Stable. The results obtained by real-time-based simulations using RTDS verify the designed control for various transient phenomena.
{"title":"Polytopic Lyapunov Function-Based Hybrid Switching Control Strategy for High Voltage Direct Current Networks","authors":"Rohan Kamat Tarcar;Marjan Popov;Aleksandra Lekić","doi":"10.1109/OAJPE.2025.3585425","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3585425","url":null,"abstract":"The increased use of High-Voltage Direct Current transmission networks requires appropriate control strategies for the converter stations, which are crucial to ensure uninterrupted energy supply. In this paper, a Polytopic Lyapunov Function-based Hybrid switching control strategy is implemented to combine the merits of Grid Following and Grid Forming control strategies by switching alternatively from one to another at the polytopes’ hyperplanes to ensure good system response even for faulty conditions. The state space equations of the control strategies are used to form the state hyperplanes for the switching rule. Since the hybrid switching control is based on the Polytopic Lyapunov Function, the system is inherently Large Signal Stable. The results obtained by real-time-based simulations using RTDS verify the designed control for various transient phenomena.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"442-454"},"PeriodicalIF":3.3,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11062912","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144606289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-16DOI: 10.1109/OAJPE.2025.3580108
Chenhan Xiao;Napoleon Costilla-Enriquez;Yang Weng
Smart grids are increasingly vulnerable to False Data Injection Attacks (FDIAs) due to their growing reliance on interconnected digital systems. Many existing FDIA techniques assume access to critical physical model information, such as grid topology, to successfully bypass Bad Data Detection (BDD). However, this assumption is often impractical, as utilities may restrict access to this data, or the evolving nature of distribution grids—particularly with the integration of renewable energy—can render this information unavailable. Current methods that address the absence of physical model lack formal guarantees for BDD evasion. To bridge this gap, we propose a novel physical-model-free FDIA framework that 1) bypasses BDD with formal guarantees and 2) maximizes the attack impact without requiring explicit physical model. Our approach leverages an autoencoder (AE) with a regularized latent space to enforce physical consistency, using historical measurements to replicate the residual error distribution, ensuring BDD evasion. Additionally, we integrate a Generative Adversarial Network (GAN) to explore the measurement manifold and induce the most significant state changes, enhancing the impact of the attack. The key innovation lies in the AE-GAN hybrid model’s ability to replicate the residual error distribution while maximizing attack efficacy, offering a performance guarantee that existing methods lack. We validate our method across 11 representative grid systems, using real power profiles simulated in MATPOWER, and demonstrate its consistent ability to bypass BDD by preserving the residual error distribution. The results highlight the robustness and generalizability of the proposed FDIA framework.
{"title":"Guaranteed False Data Injection Attack Without Physical Model","authors":"Chenhan Xiao;Napoleon Costilla-Enriquez;Yang Weng","doi":"10.1109/OAJPE.2025.3580108","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3580108","url":null,"abstract":"Smart grids are increasingly vulnerable to False Data Injection Attacks (FDIAs) due to their growing reliance on interconnected digital systems. Many existing FDIA techniques assume access to critical physical model information, such as grid topology, to successfully bypass Bad Data Detection (BDD). However, this assumption is often impractical, as utilities may restrict access to this data, or the evolving nature of distribution grids—particularly with the integration of renewable energy—can render this information unavailable. Current methods that address the absence of physical model lack formal guarantees for BDD evasion. To bridge this gap, we propose a novel physical-model-free FDIA framework that 1) bypasses BDD with formal guarantees and 2) maximizes the attack impact without requiring explicit physical model. Our approach leverages an autoencoder (AE) with a regularized latent space to enforce physical consistency, using historical measurements to replicate the residual error distribution, ensuring BDD evasion. Additionally, we integrate a Generative Adversarial Network (GAN) to explore the measurement manifold and induce the most significant state changes, enhancing the impact of the attack. The key innovation lies in the AE-GAN hybrid model’s ability to replicate the residual error distribution while maximizing attack efficacy, offering a performance guarantee that existing methods lack. We validate our method across 11 representative grid systems, using real power profiles simulated in MATPOWER, and demonstrate its consistent ability to bypass BDD by preserving the residual error distribution. The results highlight the robustness and generalizability of the proposed FDIA framework.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"429-441"},"PeriodicalIF":3.3,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11037430","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144536682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-05DOI: 10.1109/OAJPE.2025.3577012
Yexiang Chen;Charalambos Konstantinou;Daisuke Mashima;Anurag K. Srivastava;Subhash Lakshminarayana
Moving target defense (MTD) in power grids is an emerging defense technique that has gained prominence in the recent past. It aims to solve the long-standing problem of securing the power grid against stealthy attacks. The key idea behind MTD is to introduce periodic/event-triggered controlled changes to the power grid’s SCADA network/physical plant, thereby invalidating the knowledge attackers use for crafting stealthy attacks. In this paper, we provide a comprehensive overview of this topic and classify the different ways in which MTD is implemented in power grids. We further introduce the guiding principles behind the design of MTD, key performance metrics, and the associated trade-offs in MTD and identify the future development of MTD for power grid security.
{"title":"Survey of Moving Target Defense in Power Grids: Design Principles, Tradeoffs, and Future Directions","authors":"Yexiang Chen;Charalambos Konstantinou;Daisuke Mashima;Anurag K. Srivastava;Subhash Lakshminarayana","doi":"10.1109/OAJPE.2025.3577012","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3577012","url":null,"abstract":"Moving target defense (MTD) in power grids is an emerging defense technique that has gained prominence in the recent past. It aims to solve the long-standing problem of securing the power grid against stealthy attacks. The key idea behind MTD is to introduce periodic/event-triggered controlled changes to the power grid’s SCADA network/physical plant, thereby invalidating the knowledge attackers use for crafting stealthy attacks. In this paper, we provide a comprehensive overview of this topic and classify the different ways in which MTD is implemented in power grids. We further introduce the guiding principles behind the design of MTD, key performance metrics, and the associated trade-offs in MTD and identify the future development of MTD for power grid security.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"455-469"},"PeriodicalIF":3.3,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11026104","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144634859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the large-scale integration of wind power and photovoltaic (PV) into the grid, dealing with their output uncertainties and formulating more reliable scheduling strategies has become a critical challenge for the efficient operation of hydropower-dominated inter-basin hydro-wind-PV complementary systems. To quantify the uncertainty associated with wind and PV power generation, this paper proposes a method for generating wind and PV power output scenarios, combining adaptive diffusion kernel density estimation with Copula theory. Scenario reduction is then carried out using the K-means clustering algorithm. Based on this, a medium- and long-term stochastic expectation model for the inter-basin hydro-wind-PV complementary system is developed. The model is subsequently solved using the Gurobi 11.0.3 optimization solver within the MATLAB environment. A case study is conducted based on a selected inter-basin hydro-wind-PV clean energy base in China. The results demonstrate that the proposed scheduling strategy effectively addresses the unpredictability of wind and solar power, improves the overall utilization of renewable energy sources, and facilitates more efficient water level regulation at each power station. Furthermore, it significantly enhances the overall performance and efficiency of the complementary system.
{"title":"Medium- and Long-Term Optimal Stochastic Scheduling for Inter-Basin Hydro-Wind-Photovoltaic Complementary Systems Considering Wind and Solar Output Uncertainty","authors":"Chengrui Du;Yuan Gao;Lili Wang;Xiang Li;Yichen Cui;Jian Gao","doi":"10.1109/OAJPE.2025.3575734","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3575734","url":null,"abstract":"With the large-scale integration of wind power and photovoltaic (PV) into the grid, dealing with their output uncertainties and formulating more reliable scheduling strategies has become a critical challenge for the efficient operation of hydropower-dominated inter-basin hydro-wind-PV complementary systems. To quantify the uncertainty associated with wind and PV power generation, this paper proposes a method for generating wind and PV power output scenarios, combining adaptive diffusion kernel density estimation with Copula theory. Scenario reduction is then carried out using the K-means clustering algorithm. Based on this, a medium- and long-term stochastic expectation model for the inter-basin hydro-wind-PV complementary system is developed. The model is subsequently solved using the Gurobi 11.0.3 optimization solver within the MATLAB environment. A case study is conducted based on a selected inter-basin hydro-wind-PV clean energy base in China. The results demonstrate that the proposed scheduling strategy effectively addresses the unpredictability of wind and solar power, improves the overall utilization of renewable energy sources, and facilitates more efficient water level regulation at each power station. Furthermore, it significantly enhances the overall performance and efficiency of the complementary system.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"404-416"},"PeriodicalIF":3.3,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11021448","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144281282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The growing penetration of IoT devices in power grids despite its benefits, raises cybersecurity concerns. In particular, load-altering attacks (LAAs) targeting high-wattage IoT-controllable load devices pose serious risks to grid stability and disrupt electricity markets. This paper provides a comprehensive review of LAAs, highlighting the threat model, analyzing their impact on transmission and distribution networks, and the electricity market dynamics. We also review the detection and localization schemes for LAAs that employ either model-based or data-driven approaches, with some hybrid methods combining the strengths of both. Additionally, mitigation techniques are examined, focusing on both preventive measures, designed to thwart attack execution, and reactive methods, which aim to optimize responses to ongoing attacks. We look into the application of each study and highlight potential streams for future research.
{"title":"Survey of Load-Altering Attacks Against Power Grids: Attack Impact, Detection, and Mitigation","authors":"Sajjad Maleki;Shijie Pan;Subhash Lakshminarayana;Charalambos Konstantinou","doi":"10.1109/OAJPE.2025.3562052","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3562052","url":null,"abstract":"The growing penetration of IoT devices in power grids despite its benefits, raises cybersecurity concerns. In particular, load-altering attacks (LAAs) targeting high-wattage IoT-controllable load devices pose serious risks to grid stability and disrupt electricity markets. This paper provides a comprehensive review of LAAs, highlighting the threat model, analyzing their impact on transmission and distribution networks, and the electricity market dynamics. We also review the detection and localization schemes for LAAs that employ either model-based or data-driven approaches, with some hybrid methods combining the strengths of both. Additionally, mitigation techniques are examined, focusing on both preventive measures, designed to thwart attack execution, and reactive methods, which aim to optimize responses to ongoing attacks. We look into the application of each study and highlight potential streams for future research.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"220-234"},"PeriodicalIF":3.3,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10967393","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143892508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-17DOI: 10.1109/OAJPE.2025.3562091
Jon González-Ramos;Itziar Angulo;Igor Fernández;Bernhard Grasel;Alexander Gallarreta;Amaia Arrinda;David de la Vega
This paper aims to empirically characterize the long-term grid impedance variations due to Electric Vehicle Charging Processes (EVCPs) in the frequency range from 20 kHz to 500 kHz. The study is supported by a measurement campaign performed in a controlled Low Voltage (LV) grid in Austria, composed of a Secondary Substation (SS) and four houses, which statistically represents the public LV grids in Austria. The results show that different impedance states (with different spectral patterns and amplitudes) can be identified during the charging processes of all the EVs under analysis. Additionally, time variability within each impedance state is also registered. The findings, which cover the still uncharacterized frequency band from 20 kHz to 500 kHz, have important implications for the performance of Narrowband Power Line Communications (NB-PLC), the propagation of Non-Intentional Emissions (NIEs) and the definition of a reference impedance in this frequency band.
{"title":"Characterization of the Long-Term Impedance Variations Due to Electric Vehicle Charging From 20 kHz to 500 kHz","authors":"Jon González-Ramos;Itziar Angulo;Igor Fernández;Bernhard Grasel;Alexander Gallarreta;Amaia Arrinda;David de la Vega","doi":"10.1109/OAJPE.2025.3562091","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3562091","url":null,"abstract":"This paper aims to empirically characterize the long-term grid impedance variations due to Electric Vehicle Charging Processes (EVCPs) in the frequency range from 20 kHz to 500 kHz. The study is supported by a measurement campaign performed in a controlled Low Voltage (LV) grid in Austria, composed of a Secondary Substation (SS) and four houses, which statistically represents the public LV grids in Austria. The results show that different impedance states (with different spectral patterns and amplitudes) can be identified during the charging processes of all the EVs under analysis. Additionally, time variability within each impedance state is also registered. The findings, which cover the still uncharacterized frequency band from 20 kHz to 500 kHz, have important implications for the performance of Narrowband Power Line Communications (NB-PLC), the propagation of Non-Intentional Emissions (NIEs) and the definition of a reference impedance in this frequency band.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"235-244"},"PeriodicalIF":3.3,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10967552","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-15DOI: 10.1109/OAJPE.2025.3560626
Sampson E. Nwachukwu;Komla A. Folly;Kehinde O. Awodele
This paper presents a soft actor-critic (SAC)-based method for solving the solar photovoltaic (PV) Maximum Power Point Tracking (MPPT) control problem under partial shading conditions (PSCs). The MPPT method optimizes the solar PV power and ensures that it constantly operates at its “maximum power point (MPP),” regardless of the dynamics of weather conditions. Traditional MPPT methods, such as the perturb and observe (P&O) method, are commonly employed to solve the MPPT control problem. However, they often suffer from a slower convergence rate, significant oscillation near the MPP, drift problems. Additionally, in the presence of partial shading, they frequently fail to track the solar PV global maximum power point (GMPP). These problems were addressed using the deep Q-network (DQN) method. However, DQN cannot be applied to continuous action spaces. It also uses inefficient experience replay and suffers from Q-value overestimation. Thus, under PSCs and certain environmental conditions, DQN produces fluctuations of power close to the MPP or GMPP, resulting in power loss. To solve the MPPT control task, mathematical models of the Markov Decision Process, solar PV system, and boost converter were developed. Key hyperparameters affecting the SAC algorithm’s performance were also investigated. Furthermore, the P&O method was developed for comparison. Simulation results show that the SAC-based MPPT method achieved better tracking accuracy than the DQN method under standard testing conditions, varying irradiance levels, and PSCs. Also, it is shown that both the DQN and SAC methods have superior tracking performance compared to the P&O method under similar environmental conditions tested.
{"title":"Soft Actor-Critic-Based MPPT Control of Solar PV Systems Under Partial Shading Conditions","authors":"Sampson E. Nwachukwu;Komla A. Folly;Kehinde O. Awodele","doi":"10.1109/OAJPE.2025.3560626","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3560626","url":null,"abstract":"This paper presents a soft actor-critic (SAC)-based method for solving the solar photovoltaic (PV) Maximum Power Point Tracking (MPPT) control problem under partial shading conditions (PSCs). The MPPT method optimizes the solar PV power and ensures that it constantly operates at its “maximum power point (MPP),” regardless of the dynamics of weather conditions. Traditional MPPT methods, such as the perturb and observe (P&O) method, are commonly employed to solve the MPPT control problem. However, they often suffer from a slower convergence rate, significant oscillation near the MPP, drift problems. Additionally, in the presence of partial shading, they frequently fail to track the solar PV global maximum power point (GMPP). These problems were addressed using the deep Q-network (DQN) method. However, DQN cannot be applied to continuous action spaces. It also uses inefficient experience replay and suffers from Q-value overestimation. Thus, under PSCs and certain environmental conditions, DQN produces fluctuations of power close to the MPP or GMPP, resulting in power loss. To solve the MPPT control task, mathematical models of the Markov Decision Process, solar PV system, and boost converter were developed. Key hyperparameters affecting the SAC algorithm’s performance were also investigated. Furthermore, the P&O method was developed for comparison. Simulation results show that the SAC-based MPPT method achieved better tracking accuracy than the DQN method under standard testing conditions, varying irradiance levels, and PSCs. Also, it is shown that both the DQN and SAC methods have superior tracking performance compared to the P&O method under similar environmental conditions tested.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"194-208"},"PeriodicalIF":3.3,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10965705","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143892359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-14DOI: 10.1109/OAJPE.2025.3559336
Zain Ahmed;Mohsin Jamil;Ashraf Ali Khan
Multi-Energy Systems (MES) allow optimal interactions between different energy sources. Accurate load forecasting for such intricate systems would greatly enhance the performance and economic incentive to employ them. This article proposes a state-of-the-art deep learning based architecture to forecast multiple loads. The algorithm utilizes load correlations to select optimal input parameters. These optimal inputs are fed to D-TCNet (Deep – Temporal Convolution Network). This network uses multi-layer perceptrons (MLP) to encode the spatial relationship among exogenous variables which is fed to a Temporal Convolutional Network (TCN). The TCN resolves temporal information in the multi-load time series which is used for forecasting these loads for fixed output horizon. The proposed novel method is used on the energy consumption data for multi energy system of University of Austin Tempe Campus. The proposed method shows improved performance across all three energy types as well as all four seasons.
{"title":"A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting","authors":"Zain Ahmed;Mohsin Jamil;Ashraf Ali Khan","doi":"10.1109/OAJPE.2025.3559336","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3559336","url":null,"abstract":"Multi-Energy Systems (MES) allow optimal interactions between different energy sources. Accurate load forecasting for such intricate systems would greatly enhance the performance and economic incentive to employ them. This article proposes a state-of-the-art deep learning based architecture to forecast multiple loads. The algorithm utilizes load correlations to select optimal input parameters. These optimal inputs are fed to D-TCNet (Deep – Temporal Convolution Network). This network uses multi-layer perceptrons (MLP) to encode the spatial relationship among exogenous variables which is fed to a Temporal Convolutional Network (TCN). The TCN resolves temporal information in the multi-load time series which is used for forecasting these loads for fixed output horizon. The proposed novel method is used on the energy consumption data for multi energy system of University of Austin Tempe Campus. The proposed method shows improved performance across all three energy types as well as all four seasons.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"209-219"},"PeriodicalIF":3.3,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10964383","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143892433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-02DOI: 10.1109/OAJPE.2025.3556142
Hamad Alduaij;Yang Weng
Deepening the deployment of distributed energy resources requires the large-scale integration of inverter-based resources, which can deteriorate the frequency stability. Recent studies propose using neural Lyapunov-based reinforcement learning for control. While this method can be trained offline with performance guarantees, it is only optimal for specific values of system parameters, as it omits critical modeling factors like decreasing inertia and damping variation over time. To maintain the performance at varying operation points, we consider an adaptive neural Lyapunov framework that adapts the controller’s output in the presence of varying parameters. Neural networks require flexibility to maximize adaptive control performance, while stability demands monotonicity, creating an inherent conflict. In this paper, we design a partially monotonic controller that maintains stability with maximal representation capacity for parameter adaptation. Stability is ensured by having monotonicity retained for frequency while non-monotonicity is allowed for the system parameters, such as damping and inertia. The structural form of partially monotonic neural networks is used for the controller design to that end. Flexibility is allowed by the design when adaptation to changes to the system parameters is made, while the Lyapunov stability guarantee is retained. The non-monotonic layers are chosen through an adaptive layer that is designed for damping and inertia based on their relationship to control in the system equation, by which optimized output for different operating conditions is allowed.
{"title":"Reinforcement Learning for Stability-Guaranteed Adaptive Optimal Primary Frequency Control of Power Systems Using Partially Monotonic Neural Networks","authors":"Hamad Alduaij;Yang Weng","doi":"10.1109/OAJPE.2025.3556142","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3556142","url":null,"abstract":"Deepening the deployment of distributed energy resources requires the large-scale integration of inverter-based resources, which can deteriorate the frequency stability. Recent studies propose using neural Lyapunov-based reinforcement learning for control. While this method can be trained offline with performance guarantees, it is only optimal for specific values of system parameters, as it omits critical modeling factors like decreasing inertia and damping variation over time. To maintain the performance at varying operation points, we consider an adaptive neural Lyapunov framework that adapts the controller’s output in the presence of varying parameters. Neural networks require flexibility to maximize adaptive control performance, while stability demands monotonicity, creating an inherent conflict. In this paper, we design a partially monotonic controller that maintains stability with maximal representation capacity for parameter adaptation. Stability is ensured by having monotonicity retained for frequency while non-monotonicity is allowed for the system parameters, such as damping and inertia. The structural form of partially monotonic neural networks is used for the controller design to that end. Flexibility is allowed by the design when adaptation to changes to the system parameters is made, while the Lyapunov stability guarantee is retained. The non-monotonic layers are chosen through an adaptive layer that is designed for damping and inertia based on their relationship to control in the system equation, by which optimized output for different operating conditions is allowed.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"259-269"},"PeriodicalIF":3.3,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10947581","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144073416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-31DOI: 10.1109/OAJPE.2025.3556004
Koji Yamashita;Nanpeng Yu;Evangelos Farantatos;Lin Zhu
As the power transmission system’s energy sources become increasingly diversified, the grid stability is experiencing increased fluctuations, thereby necessitating more frequent and near real-time monitoring by grid operators. The power system security has been monitored through real-time contingency analysis and dynamic security assessment framework, both of which are typically based on time-domain simulations or power flow calculations. Achieving higher accuracy in grid health level prediction often requires time-consuming simulation and analysis. To improve computational efficiency, this paper develops machine learning models with phasor measurement unit (PMU) data to monitor the power system health index, focusing on rotor angle stability and frequency stability. The proposed machine learning models accurately predict frequency and angle stability indicators, essential for evaluating grid health considering various contingencies, even when dealing with limited PMU deployment in transmission grids. The proposed framework leverages a physics-informed graph convolution network and graph attention network with ordinal encoders, which are benchmarked with multi-layer perceptron models. These models are trained on dataset derived from an augmented IEEE 118-bus system with different demand levels and fuel mix, including tailored dynamic generator models, generator controller models, and grid protection models. The numerical studies explored the performance of the proposed and baseline machine learning models under both full PMU coverage and various partial PMU coverage conditions, where different data imputation methods are used for substations without PMUs. The findings from this study offer valuable insights, such as machine learning model selection and critical PMU locations regarding power equipment, into the design of data-driven grid health index prediction models for power systems.
{"title":"Graph Learning-Based Power System Health Assessment Model","authors":"Koji Yamashita;Nanpeng Yu;Evangelos Farantatos;Lin Zhu","doi":"10.1109/OAJPE.2025.3556004","DOIUrl":"https://doi.org/10.1109/OAJPE.2025.3556004","url":null,"abstract":"As the power transmission system’s energy sources become increasingly diversified, the grid stability is experiencing increased fluctuations, thereby necessitating more frequent and near real-time monitoring by grid operators. The power system security has been monitored through real-time contingency analysis and dynamic security assessment framework, both of which are typically based on time-domain simulations or power flow calculations. Achieving higher accuracy in grid health level prediction often requires time-consuming simulation and analysis. To improve computational efficiency, this paper develops machine learning models with phasor measurement unit (PMU) data to monitor the power system health index, focusing on rotor angle stability and frequency stability. The proposed machine learning models accurately predict frequency and angle stability indicators, essential for evaluating grid health considering various contingencies, even when dealing with limited PMU deployment in transmission grids. The proposed framework leverages a physics-informed graph convolution network and graph attention network with ordinal encoders, which are benchmarked with multi-layer perceptron models. These models are trained on dataset derived from an augmented IEEE 118-bus system with different demand levels and fuel mix, including tailored dynamic generator models, generator controller models, and grid protection models. The numerical studies explored the performance of the proposed and baseline machine learning models under both full PMU coverage and various partial PMU coverage conditions, where different data imputation methods are used for substations without PMUs. The findings from this study offer valuable insights, such as machine learning model selection and critical PMU locations regarding power equipment, into the design of data-driven grid health index prediction models for power systems.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"12 ","pages":"181-193"},"PeriodicalIF":3.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945887","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143800742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：