Chinese Journal of Electrical Engineering最新文献_第2页

Power Loss Optimization and Efficiency Enhancement of Si/SiC Hybrid Arms ANPC Converter Based on Particle Swarm Optimization 基于粒子群优化的Si/SiC混合臂ANPC变换器功率损耗优化及效率提升

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000111

Chao Wang;Weigen Yu;Tianlun Xia;Rulei Han;Chushan Li;Wuhua Li

{"title":"Power Loss Optimization and Efficiency Enhancement of Si/SiC Hybrid Arms ANPC Converter Based on Particle Swarm Optimization","authors":"Chao Wang;Weigen Yu;Tianlun Xia;Rulei Han;Chushan Li;Wuhua Li","doi":"10.23919/CJEE.2026.000111","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000111","url":null,"abstract":"This paper investigates power loss optimization of a Si/SiC hybrid arm active neutral point clamped (ANPC) converter. Each phase consists of a Si fundamental frequency arm, a Si low frequency arm, and a SiC high frequency arm. The Si devices process the main power at fundamental or low switching frequency, while the SiC devices process partial power and compensate the low frequency current ripple. Based on hybrid frequency modulation and coordinated current control, a particle swarm optimization (PSO) method jointly adjusts the Si arm switching frequency and two segmented power sharing ratios, <tex>$K_{1}$</tex> and <tex>$K_{2}$</tex>, according to the operating condition. Simulation results show that, when the power sharing ratio is set to 1, ripple compensation reduces the current THD from 13.4% in the Si arm to 1.34% in the total output current. At 50 kW, the best parameter combination found by PSO reduces the calculated converter loss from 442.9 W to 399.6 W, corresponding to a reduction of about 9.8%. Over the investigated power range, the PSO adjusted scheme maintains lower calculated losses than the fixed parameter scheme and also shows favorable loss performance compared with all Si and all SiC ANPC converters.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"31-46"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11740129","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219348","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}

引用次数: 0

Nodal Loss Injection Thermal Network Model for Multi-chip Junction Temperature Calculations 多芯片结温计算的节点损耗注入热网络模型

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000001

Wangyu Xie;Zhiliang Xu;Rui Wang;Xinglai Ge;Weijie Wang

{"title":"Nodal Loss Injection Thermal Network Model for Multi-chip Junction Temperature Calculations","authors":"Wangyu Xie;Zhiliang Xu;Rui Wang;Xinglai Ge;Weijie Wang","doi":"10.23919/CJEE.2026.000001","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000001","url":null,"abstract":"Accurate junction temperature calculations are essential to guarantee safe and reliable insulated-gate bipolar transistor (IGBT) operation. However, a significant thermal coupling effect exists between multiple chips, which affects the junction temperature calculation accuracy. To address this issue, a multi-chip junction temperature calculation method is proposed based on nodal loss injection. This approach is based on the thermal network model (TNM), which allows for accurate multi-chip junction temperature calculations while maintaining a small computational burden. First, a Cauer TNM system is established based on the material dimensions and thermal properties of each IGBT module layer. Second, the effects of thermal coupling and thermal loss injection on the thermal characteristics of the IGBT chips are compared. The feasibility of injecting heat loss at the nodes for thermal coupling effect equivalence is determined. Furthermore, a particle swarm optimization algorithm is utilized to determine the heat loss node injection location and injection amplitude. Accordingly, a multi-chip TNM for junction temperature calculations is developed. Finally, the proposed method accuracy for junction temperature calculations under steady and transient conditions is validated based on multi-chip parallel-connection and three-phase inverter experiments.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"278-290"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11740127","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219339","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}

引用次数: 0

Self-tuning of PI Control Parameters for Three-phase PWM Rectifiers Based on Deep Reinforcement Learning 基于深度强化学习的三相PWM整流器PI控制参数自整定

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000116

Linjun Yu;Yingchao Zhang;Jintong Nie;Tingfan Huang;Linjie Wang;Gengli Song;Shusheng Wei

{"title":"Self-tuning of PI Control Parameters for Three-phase PWM Rectifiers Based on Deep Reinforcement Learning","authors":"Linjun Yu;Yingchao Zhang;Jintong Nie;Tingfan Huang;Linjie Wang;Gengli Song;Shusheng Wei","doi":"10.23919/CJEE.2026.000116","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000116","url":null,"abstract":"The tuning and optimization of the control parameters for three-phase pulse width modulation (PWM) rectifiers play a crucial role in improving both dynamic and steady-state performance. Conventional trial-and-error methods for proportional-integral (PI) parameter tuning are time-consuming, while frequency-domain approaches (e.g., Bode plot analysis) suffer from mismatches between theoretical models and practical implementations. Therefore, an efficient and automatic parameter tuning methodology is urgently needed. An automatic tuning method is proposed for PI controller parameters of three-phase PWM rectifiers based on the twin delayed deep deterministic policy gradient (TD3) algorithm. A DRL-based two stage safe-oriented framework is established, and its effectiveness is validated through simulations in Matlab/Simulink and experiments on a DSP+FPGA-based digital control platform with hardware-in-the-loop (HIL) testing. Simulation and experimental results demonstrate that the trained agent achieves automatic parameter tuning and exhibits superior dynamic performance under load transients.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"55-65"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11740130","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219326","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}

引用次数: 0

Study on the State of Charge Estimation Method of Sodium-Ion Battery in a Wide Range of Temperatures 宽温度条件下钠离子电池充电状态估计方法研究

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2025.000117

Peng Peng;Youwei Wen;Shuai Jiang;Yongqi Li;Nan Lu;Rui Xiong

{"title":"Study on the State of Charge Estimation Method of Sodium-Ion Battery in a Wide Range of Temperatures","authors":"Peng Peng;Youwei Wen;Shuai Jiang;Yongqi Li;Nan Lu;Rui Xiong","doi":"10.23919/CJEE.2025.000117","DOIUrl":"https://doi.org/10.23919/CJEE.2025.000117","url":null,"abstract":"Sodium-ion batteries (SIBs) are superior to lithium-ion batteries (LIBs) in terms of resource sustainability and safety, making them a crucial focus of energy storage. Model-based methods for estimating the state of charge (SOC) have been widely used in LIBs. However, owing to the differences in the electrochemical characteristics of SIBs and LIBs, the applicability of these methods to SIBs should be evaluated systematically for accurate SOC estimation across a wide temperature range. (1) The behavioral characteristics of the open circuit voltage (OCV) of SIBs are analyzed across different temperatures and SOC ranges. (2) An equivalent circuit model is established, and its parameters are estimated using the forgetting factor recursive least squares algorithm. The SOC is estimated using the extended Kalman filter, unscented Kalman filter, and particle filter. (3) The similarities and differences in the OCV characteristics and SOC estimation algorithm parameter settings among SIBs, nickel manganese cobalt batteries, and lithium iron phosphate batteries are compared and discussed. The results show that all three methods are suitable for SOC estimation in SIBs under wide-temperature conditions and various operating scenarios. The root mean square and maximum absolute errors are at most 2.15% and 2.65%, respectively.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"172-183"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11739736","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219356","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}

引用次数: 0

Large Language Models in Power Systems: A Survey on Applications and Fine-Tuning 电力系统中的大型语言模型：应用与微调综述

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000040

Bing Sun;Yuan Yao;Hao Yu;Haoran Ji;Guanyu Song;Huifang Wang;Peng Li

{"title":"Large Language Models in Power Systems: A Survey on Applications and Fine-Tuning","authors":"Bing Sun;Yuan Yao;Hao Yu;Haoran Ji;Guanyu Song;Huifang Wang;Peng Li","doi":"10.23919/CJEE.2026.000040","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000040","url":null,"abstract":"Large language models (LLMs) are promising tools for knowledge representation, reasoning, and decision-making in industrial sectors. The employment of LLM in power systems is receiving increasing attention, but is still limited by the scarcity of domain-specific data, the absence of unified benchmarks, the complexity of integrating heterogeneous data, and strict physical time-varying constraints. This paper presents a focused survey on the applications and fine-tuning techniques of LLMs within power systems. The taxonomy of LLMs is first introduced, covering general foundation LLMs, industry foundation LLMs, domain-specific LLMs, and vertical scenario LLMs. The discussion then addresses the principles, categories of fine-tuning methods, evaluation metrics, and benchmark datasets that support model adaptation. Finally, the current progress, potential applications, and challenges of specialized LLMs in the power sector are examined, with emphasis on hybrid fine-tuning strategies, multimodal data integration, and deployment frameworks designed to ensure secure and reliable power system operation.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"131-152"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11739741","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219375","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}

引用次数: 0

Improving Thermal Resilience of SiC Power Converters: From Device Design, Packaging to Temperature Monitoring and Active Thermal Management 提高SiC功率转换器的热弹性：从器件设计、封装到温度监测和主动热管理

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000118

Yifei Chang;Ignacio A. Acosta;Kaichen Zhang;Antonis Stathatos;Juseong Lee;Chengmin Li

{"title":"Improving Thermal Resilience of SiC Power Converters: From Device Design, Packaging to Temperature Monitoring and Active Thermal Management","authors":"Yifei Chang;Ignacio A. Acosta;Kaichen Zhang;Antonis Stathatos;Juseong Lee;Chengmin Li","doi":"10.23919/CJEE.2026.000118","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000118","url":null,"abstract":"The rapid growth of renewable-energy systems, AI data centers, and electric transportation is driving power electronic converters toward increasingly demanding operating conditions. In particular, the widespread adoption of grid-forming converters can experience severe overcurrent stress under grid faults and transient conditions. These operating scenarios can result in substantial junction-temperature rise and accelerated degradation of SiC-based power converters, making thermal resilience a critical design consideration. In this paper, the existing approaches for enhancing thermal resilience of SiC MOSFETs based converters are comprehensively reviewed from the perspectives of power-device design, packaging, junction-temperature estimation, and (active) thermal management. We provide a holistic picture of the existing viable technologies in thermal capability enhancement of power converters, and demonstrate the multidisciplinary effort in this area. Finally, the technical challenges, research gaps, and future development trends toward highly resilient SiC power converters are identified.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"1-30"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11740136","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219378","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}

引用次数: 0

Missing Value Reconstruction Method of Renewable Energy Generation Power Based on Grayscale Mapping and Conditional Generation Adversarial Network 基于灰度映射和条件生成对抗网络的可再生能源发电功率缺失值重构方法

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2025.000156

Yang Liu;Haidong Yu;Xiaokai Zhao;Qingsong Wang

{"title":"Missing Value Reconstruction Method of Renewable Energy Generation Power Based on Grayscale Mapping and Conditional Generation Adversarial Network","authors":"Yang Liu;Haidong Yu;Xiaokai Zhao;Qingsong Wang","doi":"10.23919/CJEE.2025.000156","DOIUrl":"https://doi.org/10.23919/CJEE.2025.000156","url":null,"abstract":"With the increasingly complex construction of smart and new power grids, data loss occurs during the acquisition and storage of renewable energy power generation data. Furthermore, the rate of data loss is increasing gradually. To address this issue, a renewable energy power generation missing-value reconstruction method is proposed based on grayscale mapping and conditional generative adversarial networks (CGANs). Initially, the training dataset is preprocessed using a moving average algorithm to eliminate erroneous data, thereby enhancing the quality of the dataset. Subsequently, the one-dimensional data are transformed into a two-dimensional representation through grayscale mapping to improve the utilization of the spatiotemporal features of the data. Next, the CGAN architecture parameters are designed for multi-type data related to renewable energy generation, and adversarial training is conducted based on the loss function. Finally, based on the selection mechanism with the minimum root mean square error, the optimal sequence is selected from multiple generated sequences for missing data reconstruction. The simulation results of the numerical example show that compared with traditional methods, the method proposed in this paper has a high accuracy even in the case of a high data loss rate, with a root mean square error of approximately 3.5%.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"312-323"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11740135","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219351","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}

引用次数: 0

An Overview of Eddy-current Sensor-free Control for Bearingless Machines 无轴承电机无涡流传感器控制综述

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000013

Yu Wang;Haohui Wang;Xiaoyu Lei;Xiaolan Zhi

{"title":"An Overview of Eddy-current Sensor-free Control for Bearingless Machines","authors":"Yu Wang;Haohui Wang;Xiaoyu Lei;Xiaolan Zhi","doi":"10.23919/CJEE.2026.000013","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000013","url":null,"abstract":"Eddy-current sensor-free control (ESFC) of a bearingless motor system can reduce the volume, cost, weight, and impact of shear on transmitted liquids. Moreover, implementing an ESFC algorithm in practice is challenging due to its small displacement value and strict signal-to-noise ratio requirements. An extensive overview of ESFC for bearingless machines is provided. Initially, ESFC methods are divided into three categories, the parameter estimation method (ESFC I), observer method (ESFC II), and machine learning and intelligent algorithm (ESFC III). Different ESFC algorithms are evaluated and summarized in terms of cost, complexity, scope of application, and feasibility. This study focuses on a comparative analysis of the signal-to-noise ratio (SNR) in ESFC and speed/position sensorless control. Furthermore, an ESFC method with a high signal-to-noise ratio based on the intrinsic Hall effect is presented. Finally, the theoretical considerations and control algorithms for some existing ESFC solutions are experimentally verified, and the experimental performance of the discussed topologies is compared.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"153-171"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11739735","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219395","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}

引用次数: 0

Coordinated Control Strategy for Multi-VSGs Microgrid Based on Dynamic LADRC 基于动态LADRC的多vgs微电网协调控制策略

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000114

Heng Yang;Lizhen Wu;Wei Chen;Tingting Pei

{"title":"Coordinated Control Strategy for Multi-VSGs Microgrid Based on Dynamic LADRC","authors":"Heng Yang;Lizhen Wu;Wei Chen;Tingting Pei","doi":"10.23919/CJEE.2026.000114","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000114","url":null,"abstract":"Under the integration of large-scale distributed renewable energy resources, micro grid frequency stability is challenged by reduced rotational inertia and insufficient damping capability. To address these issues, an adaptive coordinated control strategy is proposed for multi virtual synchronous generators (VSG) micro grid based on dynamic linear active disturbance rejection control (LADRC). The primary control layer employs VSG control to provide virtual inertia support and emulate the damping characteristics of synchronous generators. Furthermore, a consensus-based dynamic LADRC secondary control scheme is developed to enhance disturbance rejection capability and achieve coordinated frequency regulation among multiple VSG units. By estimating and compensating system disturbances online, the proposed dynamic LADRC improves frequency response and damping performance without requiring an accurate system model. Subsequently, a hybrid depthwise separable convolution-transformer (DSC-transformer) network is developed to realize online prediction of LADRC observer parameters and VSG control parameters, enabling adaptive adjustment of the controller under varying operating conditions. The proposed strategy effectively improves frequency restoration, and proportional power sharing performance of multi-VSGs microgrid under load disturbances. Finally, simulation results demonstrate that the proposed strategy effectively enhances the inertia support and damping characteristics of the micro grid by accelerating active power compensation and suppressing frequency oscillations, while simultaneously achieving reliable proportional power sharing among multi-VSGs.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"94-111"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11740134","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219296","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}

引用次数: 0

Network Partitioning of Low-inertia Power Systems Considering Locational Frequency Responses under Non-Uniform Inertia Distribution 非均匀惯性分布下考虑位置频率响应的低惯量电力系统网络划分

IF 4

Chinese Journal of Electrical Engineering Pub Date : 2026-09-01 Epub Date: 2026-10-19 DOI: 10.23919/CJEE.2026.000117

Jing Wu;Zhongda Chu

{"title":"Network Partitioning of Low-inertia Power Systems Considering Locational Frequency Responses under Non-Uniform Inertia Distribution","authors":"Jing Wu;Zhongda Chu","doi":"10.23919/CJEE.2026.000117","DOIUrl":"https://doi.org/10.23919/CJEE.2026.000117","url":null,"abstract":"With the increasing replacement of synchronous generators by renewable energy resources, the equivalent inertia of modern power systems is reduced and becomes more unevenly distributed, leading to increasingly heterogeneous post-disturbance frequency responses. Conventional system-frequency or center-of-inertia (COI) descriptions capture the dominant system-wide motion but may mask locational frequency-response differences. A network partitioning method is proposed based on COI-relative local frequency deviations to identify locational frequency-response areas in low-inertia power systems. All-bus frequency trajectories are obtained using a linearized nodal frequency-response model, and trajectory-shape, derivative, and low-frequency spectral features are extracted from the COI-relative local responses. A Gaussian-kernel similarity graph is then constructed for spectral clustering. The raw number of areas is selected according to the Silhouette criterion, followed by a deterministic topological-connectivity restoration step. Case studies on the IEEE 39-bus system examine the effects of non-uniform inertia distribution, while an IEEE 118-bus study further evaluates the method under four equal-total disturbance configurations. The proposed partitions consistently exhibit stronger compactness and separation than case-matched static electrical-distance partitions, while the IEEE 118-bus results reveal both common regional structure and disturbance-dependent adaptation. These results support the use of response-oriented regionalization as a basis for preventive regional frequency-security analysis and subsequent system-level applications.","PeriodicalId":36428,"journal":{"name":"Chinese Journal of Electrical Engineering","volume":"12 3","pages":"112-130"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11739787","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219301","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}

引用次数: 0