Pub Date : 2026-02-26DOI: 10.1109/OAJPE.2026.3668297
Ahmed Mohamed;Rémy Rigo-Mariani;Vincent Debusschère
The revenues of battery energy storage systems (BESS) participating simultaneously in different markets such as energy and primary reserve has been widely investigated. In most cases, the system profitability is evaluated with optimization approaches based on historical data for prices and frequency measurements. However, in actual operations, the revenue decreases from such an ideal scenario due to uncertainties and the potential impossibility to fulfill the commitments, which translates into economic penalties. This paper proposes two-stage management strategies of a BESS participating in day-ahead and primary frequency reserve markets. The first stage consists in a day-ahead optimization of the quantities for the energy traded and capacity reserved and is based on simple forecasts. Heuristics strategies are then investigated for the real-time phase, based on actual frequency measurements at 10 seconds. Simulations are performed for data in the French market along 2021 and results obtained show that the proposed management can reach up to 90 % of the theoretical optimum profits obtained with perfect forecasts and optimal control. Especially, the real-time operation limits the penalties due to the impossibility to provide reserve when committed. Lastly, a degradation analysis of the BESS over 10 years shows that ageing remains moderated under 20 %.
{"title":"Management Strategies for a Battery Participating in Day Ahead and Primary Reserve Markets Under Uncertainties","authors":"Ahmed Mohamed;Rémy Rigo-Mariani;Vincent Debusschère","doi":"10.1109/OAJPE.2026.3668297","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3668297","url":null,"abstract":"The revenues of battery energy storage systems (BESS) participating simultaneously in different markets such as energy and primary reserve has been widely investigated. In most cases, the system profitability is evaluated with optimization approaches based on historical data for prices and frequency measurements. However, in actual operations, the revenue decreases from such an ideal scenario due to uncertainties and the potential impossibility to fulfill the commitments, which translates into economic penalties. This paper proposes two-stage management strategies of a BESS participating in day-ahead and primary frequency reserve markets. The first stage consists in a day-ahead optimization of the quantities for the energy traded and capacity reserved and is based on simple forecasts. Heuristics strategies are then investigated for the real-time phase, based on actual frequency measurements at 10 seconds. Simulations are performed for data in the French market along 2021 and results obtained show that the proposed management can reach up to 90 % of the theoretical optimum profits obtained with perfect forecasts and optimal control. Especially, the real-time operation limits the penalties due to the impossibility to provide reserve when committed. Lastly, a degradation analysis of the BESS over 10 years shows that ageing remains moderated under 20 %.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"207-216"},"PeriodicalIF":3.2,"publicationDate":"2026-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11414119","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362578","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 : 2026-02-25DOI: 10.1109/OAJPE.2026.3668128
Patrick Salter;Celina Wilkerson;Qiuhua Huang;Paulo Cesar Tabares-Velasco;Dongbo Zhao;Dmitry Ishchenko
High penetration of distributed energy resources (DER) in distribution systems, such as rooftop solar PVs, has caused voltage fluctuations which are much faster than typical voltage control devices can react to, leading to increased operation cost and reduced equipment life. Residential buildings consume about 35% of the electricity in the U.S. and are co-located with rooftop solar PV. Thus, they present an opportunity to mitigate these fluctuations locally, while benefiting both the grid and building owners. Previous works on DER-aware localized building energy management mostly focus on commercial buildings and analyzing impacts either on buildings or the grid. To fill the gaps, this paper proposes a distributed, differential predictive control scheme for residential HVAC systems for maximizing renewable local consumption while maintaining occupant comfort. In addition, a detailed controller-building-grid co-simulation platform is developed and utilized for analyzing the potential impacts of the proposed control scheme on both buildings and distribution systems. Our studies show that the proposed method can provide benefits to both the buildings’ owners and the distribution system by reducing electricity bills by 5%, voltage violations and fast fluctuations by 48%, and the number of tap changes in voltage regulators by 19%.
{"title":"Differential Predictive Control of Residential Building HVACs for Enhancing Renewable Local Consumption and Supporting Fast Voltage Control","authors":"Patrick Salter;Celina Wilkerson;Qiuhua Huang;Paulo Cesar Tabares-Velasco;Dongbo Zhao;Dmitry Ishchenko","doi":"10.1109/OAJPE.2026.3668128","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3668128","url":null,"abstract":"High penetration of distributed energy resources (DER) in distribution systems, such as rooftop solar PVs, has caused voltage fluctuations which are much faster than typical voltage control devices can react to, leading to increased operation cost and reduced equipment life. Residential buildings consume about 35% of the electricity in the U.S. and are co-located with rooftop solar PV. Thus, they present an opportunity to mitigate these fluctuations locally, while benefiting both the grid and building owners. Previous works on DER-aware localized building energy management mostly focus on commercial buildings and analyzing impacts either on buildings or the grid. To fill the gaps, this paper proposes a distributed, differential predictive control scheme for residential HVAC systems for maximizing renewable local consumption while maintaining occupant comfort. In addition, a detailed controller-building-grid co-simulation platform is developed and utilized for analyzing the potential impacts of the proposed control scheme on both buildings and distribution systems. Our studies show that the proposed method can provide benefits to both the buildings’ owners and the distribution system by reducing electricity bills by 5%, voltage violations and fast fluctuations by 48%, and the number of tap changes in voltage regulators by 19%.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"182-195"},"PeriodicalIF":3.2,"publicationDate":"2026-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11411782","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362552","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 : 2026-02-23DOI: 10.1109/OAJPE.2026.3666135
Buxin She;Ramij R. Hossain;Soumya Kundu;Marcelo A. Elizondo;Veronica Adetola
Hybrid DC-AC power systems integrating inverter-based resources (IBRs) and multi-terminal high-voltage direct current (MTDC) networks offer a promising paradigm for future power grids, while introducing challenges for modeling, stability analysis, and control design. This paper develops a hybrid symbolic-numerical modeling framework and tool to characterize the parametric small-signal stability of DC-AC coupled power systems. The proposed approach constructs parametric state-space models to enable efficient representation of system dynamics under varying control parameters and network configurations, with target parameters retained as symbolic variables and the remainder treated numerically. The stability analysis framework covers eigenvalue, sensitivity, and stability boundary and region characterization. Enhanced linear matrix inequality (LMI) techniques are proposed to directly certify small-signal stability over regions of parameter space while also reducing the conservativeness and computational burden. The resulting tools and frameworks enable rapid parametric model construction across diverse grid conditions, thereby facilitating stability-informed control and operation in DC–AC power systems.
{"title":"Hybrid Symbolic-Numerical Modeling and Parametric Stability Analysis of DC–AC Power Systems","authors":"Buxin She;Ramij R. Hossain;Soumya Kundu;Marcelo A. Elizondo;Veronica Adetola","doi":"10.1109/OAJPE.2026.3666135","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3666135","url":null,"abstract":"Hybrid DC-AC power systems integrating inverter-based resources (IBRs) and multi-terminal high-voltage direct current (MTDC) networks offer a promising paradigm for future power grids, while introducing challenges for modeling, stability analysis, and control design. This paper develops a hybrid symbolic-numerical modeling framework and tool to characterize the parametric small-signal stability of DC-AC coupled power systems. The proposed approach constructs parametric state-space models to enable efficient representation of system dynamics under varying control parameters and network configurations, with target parameters retained as symbolic variables and the remainder treated numerically. The stability analysis framework covers eigenvalue, sensitivity, and stability boundary and region characterization. Enhanced linear matrix inequality (LMI) techniques are proposed to directly certify small-signal stability over regions of parameter space while also reducing the conservativeness and computational burden. The resulting tools and frameworks enable rapid parametric model construction across diverse grid conditions, thereby facilitating stability-informed control and operation in DC–AC power systems.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"145-156"},"PeriodicalIF":3.2,"publicationDate":"2026-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11407964","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299744","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 : 2026-02-23DOI: 10.1109/OAJPE.2026.3666455
Sunil Subedi;Jongchan Choi;Yaosuo Xue
Traditionally, distribution system planning has focused on steady-state analyses, with limited consideration of dynamic behavior. However, as large or medium-scale inverter-based resources (IBRs), particularly grid-following (GFL) inverters in commercial or industry buildings, become more prevalent, understanding their dynamic impact is essential for grid planning and operation. This article presents an innovative deep-learning (DL)-approach using convolutional neural networks technique to model the GFL inverters. Developed from real grid-tied commercial IBR transient data, these dynamic DL models overcome proprietary constraints by requiring minimal knowledge of internal converter physics while maintaining high accuracy and flexibility. To demonstrate their applicability, the models were incorporated into GridLAB-D, an open-source, three-phase distribution analysis tool. This integration enables dynamic simulations of large-scale distribution networks with high IBR penetration stability analysis. Rigorous testing and validation, aligned with industry standards, confirmed the reliability and efficiency of this approach, paving the way for enhanced planning and operational assessments of modern power systems.
{"title":"Dynamic Validation of CNN-Based Surrogate Models for Inverter-Based Resources in Open-Source Solvers","authors":"Sunil Subedi;Jongchan Choi;Yaosuo Xue","doi":"10.1109/OAJPE.2026.3666455","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3666455","url":null,"abstract":"Traditionally, distribution system planning has focused on steady-state analyses, with limited consideration of dynamic behavior. However, as large or medium-scale inverter-based resources (IBRs), particularly grid-following (GFL) inverters in commercial or industry buildings, become more prevalent, understanding their dynamic impact is essential for grid planning and operation. This article presents an innovative deep-learning (DL)-approach using convolutional neural networks technique to model the GFL inverters. Developed from real grid-tied commercial IBR transient data, these dynamic DL models overcome proprietary constraints by requiring minimal knowledge of internal converter physics while maintaining high accuracy and flexibility. To demonstrate their applicability, the models were incorporated into GridLAB-D, an open-source, three-phase distribution analysis tool. This integration enables dynamic simulations of large-scale distribution networks with high IBR penetration stability analysis. Rigorous testing and validation, aligned with industry standards, confirmed the reliability and efficiency of this approach, paving the way for enhanced planning and operational assessments of modern power systems.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"157-168"},"PeriodicalIF":3.2,"publicationDate":"2026-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11407947","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299690","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}
Voltage control in modern power systems has become increasingly complex due to the high penetration of renewable generation. Numerous solutions have been proposed from both the transmission and distribution sides, involving generators and system operators. However, the contribution of loads has remained limited, mainly to demand shifting and basic demand response strategies. This work introduces a novel approach that leverages digital twins to enhance the active participation of loads in supporting voltage control. Unlike traditional methods, the proposed framework builds digital twins exclusively from measurable data, enabling virtually any converter-interfaced load connected to a grid, regardless of whether the network is fully known or not, to contribute effectively to voltage regulation. The methodology is first demonstrated through a parametric study, which evaluates the impact of different load behaviors and control strategies on network voltage stability. To further validate the approach, hardware-in-the-loop (HIL) experiments are conducted, confirming the feasibility of real-time implementation. Four voltage control use-cases are developed and tested for a controllable thermal load, showing that even individual loads can provide meaningful support to grid voltage regulation. The results highlight the potential of data-driven digital twins to unlock new, scalable, and flexible contributions from loads, reinforcing the stability of future power systems with high renewable penetration.
{"title":"A Measurement-Driven Digital-Twin Methodology for Flexible Loads Voltage Control in Unknown Grids","authors":"Jesús Araúz;Antoine Labonne;Yvon Besanger;Fréderic Wurtz;Simon Waczowicz;Veit Hagenmeyer","doi":"10.1109/OAJPE.2026.3666226","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3666226","url":null,"abstract":"Voltage control in modern power systems has become increasingly complex due to the high penetration of renewable generation. Numerous solutions have been proposed from both the transmission and distribution sides, involving generators and system operators. However, the contribution of loads has remained limited, mainly to demand shifting and basic demand response strategies. This work introduces a novel approach that leverages digital twins to enhance the active participation of loads in supporting voltage control. Unlike traditional methods, the proposed framework builds digital twins exclusively from measurable data, enabling virtually any converter-interfaced load connected to a grid, regardless of whether the network is fully known or not, to contribute effectively to voltage regulation. The methodology is first demonstrated through a parametric study, which evaluates the impact of different load behaviors and control strategies on network voltage stability. To further validate the approach, hardware-in-the-loop (HIL) experiments are conducted, confirming the feasibility of real-time implementation. Four voltage control use-cases are developed and tested for a controllable thermal load, showing that even individual loads can provide meaningful support to grid voltage regulation. The results highlight the potential of data-driven digital twins to unlock new, scalable, and flexible contributions from loads, reinforcing the stability of future power systems with high renewable penetration.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"196-206"},"PeriodicalIF":3.2,"publicationDate":"2026-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11400545","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362553","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 : 2026-02-18DOI: 10.1109/OAJPE.2026.3666003
Yuri O. Cota;Mariana Netto;Gilney Damm;Jefferson S. Costa;Ekom E. Okpo;Alfeu J. Sguarezi Filho
The growing penetration of wind generation, particularly systems based on Doubly-Fed Induction Generators (DFIG), requires advanced control strategies to ensure stable and reliable operation under adverse conditions such as distorted voltages and parametric variations. Finite Control Set Model Predictive Control (FCS-MPC) has emerged as a promising solution, but the lack of formal stability guarantees hinders its practical adoption. This paper addresses this gap by proposing a rigorous FCS-MPC framework for rotor current control of DFIGs, whose exponential stability is formally proven using Lyapunov theory. Experimental validation demonstrates high-performance operation, achieving steady-state tracking errors below 2%, settling times under 1 ms, negligible overshoot (<1%), and Total Harmonic Distortion (THD) within 5%, even under distorted grid voltages. Comparative analysis with conventional MPC and Deadbeat control highlights the superiority and robustness of the proposed approach, establishing it as an effective solution for modern wind energy conversion systems.
{"title":"Lyapunov-Based of Finite Control Set Applied to DFIG Operating Under Distorted Voltage and Parametric Uncertainties","authors":"Yuri O. Cota;Mariana Netto;Gilney Damm;Jefferson S. Costa;Ekom E. Okpo;Alfeu J. Sguarezi Filho","doi":"10.1109/OAJPE.2026.3666003","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3666003","url":null,"abstract":"The growing penetration of wind generation, particularly systems based on Doubly-Fed Induction Generators (DFIG), requires advanced control strategies to ensure stable and reliable operation under adverse conditions such as distorted voltages and parametric variations. Finite Control Set Model Predictive Control (FCS-MPC) has emerged as a promising solution, but the lack of formal stability guarantees hinders its practical adoption. This paper addresses this gap by proposing a rigorous FCS-MPC framework for rotor current control of DFIGs, whose exponential stability is formally proven using Lyapunov theory. Experimental validation demonstrates high-performance operation, achieving steady-state tracking errors below 2%, settling times under 1 ms, negligible overshoot (<1%), and Total Harmonic Distortion (THD) within 5%, even under distorted grid voltages. Comparative analysis with conventional MPC and Deadbeat control highlights the superiority and robustness of the proposed approach, establishing it as an effective solution for modern wind energy conversion systems.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"169-181"},"PeriodicalIF":3.2,"publicationDate":"2026-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11398100","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362563","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 : 2026-02-12DOI: 10.1109/OAJPE.2026.3651003
{"title":"IEEE Open Access Journal of Power and Energy Publication Information","authors":"","doi":"10.1109/OAJPE.2026.3651003","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3651003","url":null,"abstract":"","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"C2-C2"},"PeriodicalIF":3.2,"publicationDate":"2026-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11394821","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146176012","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 : 2026-02-12DOI: 10.1109/OAJPE.2026.3651145
{"title":"Information for authors","authors":"","doi":"10.1109/OAJPE.2026.3651145","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3651145","url":null,"abstract":"","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"C3-C3"},"PeriodicalIF":3.2,"publicationDate":"2026-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11394822","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175915","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 : 2026-01-30DOI: 10.1109/OAJPE.2026.3659790
Jared Paull;Walid Hatahet;Liwei Wang;Wei Li
Electromagnetic transient (EMT) simulation of power electronic converters is critical for analysis, design, and fast control prototyping of power and energy systems. This paper proposes a multi-granular GPU parallel-rate exponential integrator (EI) algorithm for fast offline EMT simulation of power electronic systems. The proposed parallel-rate EI algorithm utilizes the massively parallel GPU architecture to compute multiple discretization steps in parallel. The matrix-vector computations of the EI algorithm within each time step are also parallelized. Additionally, a novel GPU-based framework is proposed for numerically efficient precomputation of matrix exponentials before a simulation loop starts. The high degree of parallelism leads to large simulation speedups compared to single-thread CPU implementations. The discretization technique of high-order EI algorithm is absolutely stable with no numerical ringing and can achieve accurate differential equation discretization with large time-step sizes. The proposed parallel-rate EI solver is further applied to detect passive/diode switching events accurately. Two case studies are used to demonstrate the accuracy and efficiency of the parallel-rate EI algorithm. Two additional case studies showcase the benefit of the proposed parallel precomputation technique for matrix exponentials.
{"title":"GPU Parallel-Rate Exponential Integrator Algorithm for Efficient Simulation of Power Electronic Systems","authors":"Jared Paull;Walid Hatahet;Liwei Wang;Wei Li","doi":"10.1109/OAJPE.2026.3659790","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3659790","url":null,"abstract":"Electromagnetic transient (EMT) simulation of power electronic converters is critical for analysis, design, and fast control prototyping of power and energy systems. This paper proposes a multi-granular GPU parallel-rate exponential integrator (EI) algorithm for fast offline EMT simulation of power electronic systems. The proposed parallel-rate EI algorithm utilizes the massively parallel GPU architecture to compute multiple discretization steps in parallel. The matrix-vector computations of the EI algorithm within each time step are also parallelized. Additionally, a novel GPU-based framework is proposed for numerically efficient precomputation of matrix exponentials before a simulation loop starts. The high degree of parallelism leads to large simulation speedups compared to single-thread CPU implementations. The discretization technique of high-order EI algorithm is absolutely stable with no numerical ringing and can achieve accurate differential equation discretization with large time-step sizes. The proposed parallel-rate EI solver is further applied to detect passive/diode switching events accurately. Two case studies are used to demonstrate the accuracy and efficiency of the parallel-rate EI algorithm. Two additional case studies showcase the benefit of the proposed parallel precomputation technique for matrix exponentials.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"121-134"},"PeriodicalIF":3.2,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11368877","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175914","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 : 2026-01-22DOI: 10.1109/OAJPE.2026.3657191
Jian Huang;Yongli Zhu;Linna Xu;Zhe Zheng;Wenpeng Cui;Mingyang Sun
To ensure a resilient and autonomous execution of the photovoltaic power forecasting task for a remote microgrid in cloud-less and weak-communication situations, an original study regarding edge-side model-training is conducted on a resource-constrained smart meter. On-device training of two representative machine learning models is investigated: a gradient boosting tree model and a recurrent neural network model. Besides, to speed up the training process of the neural networks on the smart meter, “reduced”- and “mixed”-precision training schemes are also devised, which can achieve about 2X speed-up. Experiments on a real dataset demonstrate the feasibility of economically achieving grid-edge intelligence via the existing metering infrastructures.
{"title":"On-Device Training of PV Power Forecasting Models in a Smart Meter for Grid Edge Intelligence","authors":"Jian Huang;Yongli Zhu;Linna Xu;Zhe Zheng;Wenpeng Cui;Mingyang Sun","doi":"10.1109/OAJPE.2026.3657191","DOIUrl":"https://doi.org/10.1109/OAJPE.2026.3657191","url":null,"abstract":"To ensure a resilient and autonomous execution of the photovoltaic power forecasting task for a remote microgrid in cloud-less and weak-communication situations, an original study regarding edge-side model-training is conducted on a resource-constrained smart meter. On-device training of two representative machine learning models is investigated: a gradient boosting tree model and a recurrent neural network model. Besides, to speed up the training process of the neural networks on the smart meter, “reduced”- and “mixed”-precision training schemes are also devised, which can achieve about 2X speed-up. Experiments on a real dataset demonstrate the feasibility of economically achieving grid-edge intelligence via the existing metering infrastructures.","PeriodicalId":56187,"journal":{"name":"IEEE Open Access Journal of Power and Energy","volume":"13 ","pages":"116-120"},"PeriodicalIF":3.2,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11361107","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146176026","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：