Xiaoyu Zhang;Qiuye Sun;Tianyi Li;Yumeng Song;Zhongming Yao;Yushuai Li
{"title":"A Survey on Large Language Models Enhanced Reinforcement Learning for Smart Grid","authors":"Xiaoyu Zhang;Qiuye Sun;Tianyi Li;Yumeng Song;Zhongming Yao;Yushuai Li","doi":"10.35833/MPCE.2025.000685","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000685","url":null,"abstract":"The smart grid (SG) is a modern power system that leverages digital technologies to enhance the generation, delivery, and consumption of electric power. Reinforcement learning (RL) plays an important role in SG by helping make smart decisions. However, RL faces challenges such as sparse rewards, poor generalization, and difficulty in representing and interpreting regulatory constraints. Large language models (LLMs) offer new opportunities to address these challenges by understanding natural language, leveraging external knowledge, and enhancing reasoning. This paper presents a comprehensive review of LLM-enhanced RL for SG. It first analyzes the key challenges of RL and introduces how LLMs help enhance the performance, including a taxonomy based on the integration patterns and functions. It then reviews applications of LLM-enhanced RL for SG, with a focus on energy management, operational control, electricity market, and hardware design. Finally, it discusses research challenges and future directions of LLM-enhanced RL for SG. This paper aims to provide a clear framework for researchers to apply LLMs in RL and to promote the application of RL for SG.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1549-1567"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11458071","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Critical Boundary Index-Based Preventive and Post-Detection Defense Approach Against False Data Injection Attacks in Power Systems with Distributed Energy Resources","authors":"Masahiro Furukakoi;Akito Nakadomari;Akie Uehara;Paras Mandal;Mitsunaga Kinjo;Tomonobu Senjyu","doi":"10.35833/MPCE.2025.000962","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000962","url":null,"abstract":"Cybersecurity in power systems with distributed energy resources (DERs) has become a serious challenge. This paper proposes a critical boundary index (CBI) -based preventive and post-detection defense approach against false data injection attacks (FDIAs) in power systems with DERs. The proposed approach utilizes the inherent attack resistance of CBI, a voltage stability index previously developed by the authors, to both preventively limit the impact of attacks and provide high-sensitivity post-attack detection of data manipulation. This paper quantitatively evaluates voltage stability monitoring approaches against FDIAs, comparing the resilience of CBI with conventional indices. Verification is conducted using IEEE 5-bus and 118-bus test systems with two types of FDIA scenarios: detection-avoidance type and misdirection type. The results demonstrate that when CBI is employed for monitoring, the achievable false data injection by attackers aiming to cause voltage drops is significantly reduced compared with conventional indices, maintaining system voltage within stable ranges. Additionally, CBI shows 1.5-2 times higher detection sensitivity compared with conventional indices. Notably, CBI demonstrates effective detection capability even against sophisticated attacks that falsely show improvement in voltage stability indices, which are difficult to detect using conventional approaches. These results confirm that the proposed approach effectively constrains attacker capabilitieswhile providing enhanced detection sensitivity. Based on these findings, the proposed approach is shown to provide high sensitivity to even minor data tampering, offering a multi-layered defense combining prevention and detection for power systems with DERs.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1744-1755"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11554476","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149033590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Coordinated Expansion Planning for Multi-Infeed LCC-HVDC Systems Considering Generalized Short-Circuit Ratio Constraints","authors":"Liang Shao;Xiaoru Zhang;Yusheng Xue;Zongqiang Zheng;Feng Xue;Fushuan Wen","doi":"10.35833/MPCE.2025.000669","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000669","url":null,"abstract":"Line-commutated converter based high-voltage direct current (LCC-HVDC) systems may diminish the power grid strength, potentially precipitating static voltage stability issues. To maintain the required level of power grid strength while minimizing the total system cost, this paper proposes a coordinated expansion planning framework for multi-infeed LCC-HVDC (MI-HVDC) systems that explicitly incorporates generalized short-circuit ratio (gSCR) constraints. Specifically, the gSCR requirement is innovatively formulated as a semidefinite constraint, and the proposed framework is reformulated as a mixed-integer semidefinite programming (MISDP) model to guarantee the global optimality. Then, to address the computational intractability of the MISDP model, a generalized Benders decomposition (GBD)-based approach is employed for efficient solution. Time-domain simulations conducted in PSCAD/EMT-DC demonstrate the effectiveness of the proposed framework in enhancing the voltage stability of MI-HVDC systems.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1921-1932"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11471008","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hong Yu;Yong Zhao;Manli Yan;Yuanzheng Li;Yaowen Yu
{"title":"Data-Model Hybrid-Driven Approach for Unit Commitment Considering Conditional Transmission Section Limits","authors":"Hong Yu;Yong Zhao;Manli Yan;Yuanzheng Li;Yaowen Yu","doi":"10.35833/MPCE.2025.000658","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000658","url":null,"abstract":"Conditional transmission section limits (C-TSLs) depend on the actual operational state of the power grid, such as commitment status and reserve capacity, exacerbating the computational complexity of the unit commitment problem. To overcome this complexity, this paper proposes a data-model hybrid-driven approach for solving unit commitment problems. Furthermore, a unit commitment prediction algorithm is proposed to reduce the problem scale by fixing a subset of unit commitment variables, thereby eliminating inactive C-TSL intervals. Specifically, the proposed algorithm incorporates a convolutional attention module to perform deep feature extraction and enhancement on time series data, complemented by a multi-head cross-attention mechanism designed to synthesize temporal and non-temporal features. The introduction of attention mechanisms enhances the prediction performance of the proposed algorithm. Numerical results demonstrate that the proposed approach significantly reduces the number of variables and constraints in the unit commitment model, thereby expediting the solution process while maintaining high solution accuracy.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1708-1719"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11473943","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yanhong Liu;Xinfei Yan;Haiwang Zhong;Chongqing Kang
{"title":"Fast Infeasibility Analysis of Network-constrained Unit Commitment Model","authors":"Yanhong Liu;Xinfei Yan;Haiwang Zhong;Chongqing Kang","doi":"10.35833/MPCE.2025.000736","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000736","url":null,"abstract":"With the rapid development of the electricity market and the increasingly stringent operational standards, the longstanding infeasible network-constrained unit commitment (NCUC) issue in market operations is calling for effective and systematic methods that can analyze and repair infeasible NCUC models. Studies show that a set of irreducible infeasible subsets (IISs) may lead to infeasibility in a model. The identification and repair of IISs can correct the original infeasible model, while traditional filtering algorithms for IIS location often exhibit computational inefficiency. A fast infeasibility analysis framework of NCUC and a relaxation filtering algorithm are proposed in this paper to analyze infeasible NCUC models more comprehensively and effectively. The infeasibility analysis framework employs subsystem-based IIS analysis method, generating and analyzing simplified models to repair the infeasible NCUC. The relaxation filtering algorithm can achieve greater effectiveness of IIS identification in the holistic NCUC compared with traditional general filtering algorithms. The case studies show that the proposed framework and algorithm can analyze and repair infeasible NCUC models effectively, reducing the analytical time by at least one order of magnitude compared with that of traditional general methods.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1696-1707"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11417459","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optimal Scheduling of Data Center Virtual Power Plant in Electricity–Carbon Joint Market Under Uncertainty","authors":"Ziwei Zhao;Chen Yang;Chen Liang;Dan Xu;Yilin Zhang;Junjie Tang","doi":"10.35833/MPCE.2025.000414","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000414","url":null,"abstract":"As an emerging demand response (DR) resource, data centers (DCs) have garnered significant attention due to their ability to adapt to stochastic renewable energy fluctuations. To fully exploit the price signals from diverse markets and achieve the flexible complementarity between DCs and virtual power plant (VPP) resources, this paper proposes a coordinated framework for the data center virtual power plant (DCVPP) participating in the electricity-carbon joint market under uncertainty. Specifically, a refined model of power consumption for cooling system in DC is developed based on the second-order equivalent thermal parameter (ETP) theory, which enhances the accuracy of load modeling and maximizes the potential for DR. To address the penalty risks associated with forecast uncertainties in renewable energy generation and load demands, a risk-based reserve scheme is developed by integrating the Gaussian copula with two-sided superquantile theory. This scheme provides a coordinated mechanism for quantifying the correlations between forecast errors. On this basis, an optimal scheduling strategy of DCVPPs is formulated in the electricity-carbon joint market, which coordinates DC load flexibility with distributed energy resources to optimize economic performance and reduce carbon emissions. Case studies demonstrate that the proposed refined model achieves an root mean square error (RMSE) reduction ranging from 57.9% to 68.8%, and the risk-based reserve scheme reduces penalty costs by 24.4%. Furthermore, the integration of the carbon trading mechanism in the proposed optimal scheduling strategy results in an increase of <tex>${$}$</tex>259.9 in economic benefits and an average reduction of 14% in carbon emissions.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1896-1908"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11458066","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149033604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Robust Stability Analysis and Design of Parallel Grid-Forming Converters in Islanded Microgrid","authors":"Xin Jin;Zhipeng Zhou;Ningyi Dai","doi":"10.35833/MPCE.2025.000833","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000833","url":null,"abstract":"The stability of microgrid (MG) is affected by the interaction between parallel inverter-based resources (IBRs), especially when grid-forming (GFM) control is implemented inis-landed MG to provide voltage support. In this paper, the robust stability analysis using structured singular value (SSV) approach and design of parallel GFM converters in islanded MG by structured <tex>$mu$</tex> synthesis are presented, catering to its multi-input multi-output (MIMO) nature. Specifically, a modular state-space model of parallel GFM converters in islanded MG is derived with component connection method (CCM). Model uncertainty is captured against the nominal model, delivering a certain degree of robustness for stability analysis, in the face of potential parametric uncertainty and resonance effects induced by parallel GFM converters. Moreover, the effects of system parameters on robust stability are comprehensively analyzed, and a robust design is proposed for virtual impedance (VI) tuning. The accuracy of modelling and robust stability analysis, as well as the effectiveness of the robust design is validated in MATLAB/Simulink/PLECS and real-time digital simulator (RTDS).","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1672-1683"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11471006","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149033846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Transient Stability Enhancement of Grid-forming Inverter-based Resources by Optimizing Angle of Saturated Current","authors":"Ali Arjomandi-Nezhad;Bikash C. Pal","doi":"10.35833/MPCE.2024.001292","DOIUrl":"https://doi.org/10.35833/MPCE.2024.001292","url":null,"abstract":"The current saturation challenges the transient stability of grid-forming (GFM) inverter-based resources (IBRs) during large disturbances by introducing the current-saturated stable equilibrium point (CS-SEP). Convergence into the post-fault CS-SEP is an undesired post-disturbance scenario. The angle of the saturated current significantly affects the post-disturbance trajectory. Moreover, the voltage immediately after the converter returns to the normal operation mode depends on this angle. If the difference between this voltage and the reference voltage is large, huge power oscillations occur superim-posed on the second-order power swing. The voltage-error-induced power oscillation appears as a disturbance on the active power control loop and affects transient stability. In this paper, the angle of the saturated current is adjusted to minimize the voltage error immediately after the GFM IBR returns to the normal operation mode and prevent the convergence into CS-SEP. To do so, a closed-form expression for the angle of the saturated current at which the voltage transient is minimized is derived in the first stage. Then, an optimization is formulated to minimize the deviation of the angle of the saturated current from the angle calculated in the first stage while ensuring that CS-SEP convergence is avoided. This optimal control method, which enhances transient stability by adaptively tuning the angle of the saturated current, is validated through simulation.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1647-1658"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318101","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149033910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis","authors":"Xiangjun Zeng;Xiangqing Fang;Binqiao Zhang;Chen Feng;Shengyuan Zhou","doi":"10.35833/MPCE.2025.000413","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000413","url":null,"abstract":"Wind turbine (WT) fault diagnosis using supervisory control and data acquisition (SCADA) data is challenged by severe class imbalance, which deteriorates the recognition of minority class. To address this issue, a spatiotemporal feature extraction framework from integrated multivariate time series (IMTS) is proposed, with coordinated designs in data representation, model architecture, and loss regularization. First, a dual-label IMTS is constructed to fuse multi-source SCADA data via sliding windows. Second, a spatiotemporal feature learning architecture is developed by integrating a multi-scale convolutional neural network (MCNN) and a stacked long short-term memory (LSTM) network and designing the MCNN-LSTM model. Finally, an improved weighted cross-entropy loss is proposed, which incorporates kernel density estimation-based sample probabilities as regularization, thereby directing model attention toward minority and hard-to-classify samples. Experimental results on a real SCADA dataset demonstrate that the proposed MCNN-LSTM model achieves a Macro-recall score of 0.943 and a G-mean score of 0.942, outperforming comparison models in comprehensive performance.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1768-1779"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11417460","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Convex Hull Method for Real-Time Pricing in Smart Grid with Multiple Generating Units via Smoothing Technique","authors":"Zeyi Zhu;Yan Gao;Xiaodong Ding;Youmeng He","doi":"10.35833/MPCE.2025.000688","DOIUrl":"https://doi.org/10.35833/MPCE.2025.000688","url":null,"abstract":"The increasing electricity demand imposes higher requirements on the power system to generate electricity and balance the supply and demand. The unit commitment problem is critical in power systems for determining the generation schedules and market prices. However, its inherent nonconvexity and discontinuity pose significant challenges to efficient pricing. The convex hull method addresses this nonconvexity by taking the slope of the convex envelope of cost function for generating units over the convex hull of the feasible set as the price. Nevertheless, due to the nonsmoothness, the convex hull price cannot be computed by using the gradient-based methods, which increases the computational complexity. Smoothing techniques provide an effective way by converting the nonsmooth problem into a smooth one, which enhances the computational efficiency. This paper proposes a real-time pricing scheme for the smart grid that incorporates renewable energy generation and energy storage devices on the demand side and multiple generating units on the supply side. Considering the dynamic demand of users and startup costs of generating units, a social welfare maximization model is formulated. By employing the convex hull method, the convex envelopes of the cost functions for generating units are derived. The smoothing technique is then employed to convert the original model into a smooth one. Furthermore, a distributed iterative algorithm on the basis of gradient projection method is developed by leveraging the separable structure of the variables to solve the model efficiently. Simulation results validate the feasibility and effectiveness of the convex hull method for real-time pricing in smart grid with multiple generating units via smoothing technique.","PeriodicalId":51326,"journal":{"name":"Journal of Modern Power Systems and Clean Energy","volume":"14 5","pages":"1909-1920"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11473945","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}