Volume 41, Issue 4 (2026)
Smart grid
Mechanism-guided temporal representation and risk-aware identification method for high-impedance ground faults in distribution networks under multiple types of confusing disturbances
Huaizhou Liu, Junqin Liu, and Feng Deng
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.001
To address the difficulty of representing the weak features of high-impedance ground faults （HIFs ） under multiple types of confusing transient disturbances，as well as the insufficient consideration of missed-detection risk，a mechanism-guided identification method based on multi-scale residual attention temporal representation and a risk-aware light gradient boosting machine （LightGBM ） is proposed.First，using the original zero-sequence current as input，a multi-scale residual attention temporal convolutional network is constructed based on the HIF waveform mechanisms of random arcing，local spikes，half-cycle asymmetry，and nonstationary fluctuations to adaptively extract deep discriminative features at different time scales.Second，the deep temporal features are input into LightGBM，and protection-oriented class risk weights are introduced so that the classification boundary is adjusted toward reducing HIF missed detections.Finally，simulation and a 10 kV full-scale test are conducted to verify the effectiveness of the proposed method.The results show that the proposed method achieves an overall identification accuracy of 98.19% and an HIF recall of 99.17% on the simulation validation set.On the independent test set from the full-scale test，the overall identification accuracy and HIF recall are 97.69% and 98.00%，respectively. The proposed method effectively reduces the risk of HIF missed detection under multiple types of confusing disturbances.
Vulnerability assessment of power grid nodes based on MOO-GCN
Kang Xing, Ziqing Zhang, Qunmin Yan, and Jingjing Tian
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.002
The continued expansion and increasing structural complexity of new power systems have made the limitations of traditional node vulnerability analysis methods increasingly evident in terms of computational efficiency，nonlinear feature analysis capability，and assessment accuracy.To address this issue，a power grid node vulnerability assessment method based on a multi-objective optimization graph convolutional network （MOO-GCN ） was first proposed.Then，a multi-objective optimization mechanism was introduced into the traditional GCN framework to construct a three-component joint loss function integrating reconstruction loss，neighborhood embedding loss，and ranking consistency loss，thus achieving comprehensive assessment through dynamic weight adjustment.Finally，the IEEE 300-node，IEEE 500-node，and IEEE 1354 -node distribution networks were used as test cases，and the model performance was comprehensively evaluated using the TOP-K Jaccard similarity，Kendall ’s tau rank correlation coefficient，and the SI spreading model，thus verifying the feasibility and effectiveness of the proposed method.The results show that the MOO-GCN-based power grid node vulnerability assessment method outperforms both traditional methods and the GCN-based method in terms of infection speed and ranking consistency，and that the proposed algorithm improves the stability and accuracy of node ranking.
Wide-area optimal control method for synchronous generators to enhance transient voltage stability of receiving-end grids
Peng Wang, Chenyi Zheng, Qing Wang, and Yi Tang
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.003
Synchronous generators are important reactive power sources in DC receiving-end grids.However，the traditional automatic excitation voltage regulation method，which controls reactive power output solely according to the voltage drop at the generator terminal，fails to fully exploit the reactive power and voltage regulation potential of synchronous generators.To address this issue，a wide-area optimal control framework and method for synchronous generators are proposed to enhance the transient voltage stability of receiving-end grids.First，fault scanning was conducted based on typical operating conditions，and voltage-weak nodes in the receiving-end grid were identified according to the nodal voltage-drop time-area and transient voltage stability margin.Second，the reactive power support capability of each synchronous generator for the voltage-weak nodes was evaluated.For synchronous generators with strong support capability，a remote voltage regulation control unit was added to the excitation system，and the voltage signals of the weak nodes were introduced to form supplementary control.Third，considering the overvoltage constraints of the controlled synchronous generators and nearby renewable energy sources，the key parameters of the remote voltage regulation control units were optimized with the objective of minimizing the voltage-drop time-area of the weak nodes.Finally，simulations were conducted based on data from a simple system and an actual power grid.The results show that the proposed method not only avoids overvoltage but also effectively enhances the transient voltage stability of the power grid.
Joint identification method of distribution network topology and line parameters based on sparse regression and clustering correction
Ruifeng Zhang, Bo Li, Kai Liao, Xueshun Ye, Jun Zhou, and Lei Wu
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.004
The inconsistency between equipment records and actual topology/line parameters during the long-term operation of distribution networks seriously limits the safe and efficient operation of distribution networks.To this end，a joint identification method for distribution network topology and line parameters based on sparse regression and clustering correction is proposed.Firstly，the coupling relationship between measurement data and distribution network topology/line parameters is analyzed to build the joint identification model.Subsequently，a preliminary topology identification-line parameter calculation-joint precision identification method is introduced to pre-identify topology by employing voltage similarity.Then，line parameters are solved via sequential backward selection and sparse regression，and accurate topology is extracted through an improved spatial clustering algorithm to achieve joint identification of distribution topology and line parameters.Finally，simulation experiments are carried out to verify the feasibility and effectiveness of the proposed method.Simulation experiments demonstrate that the proposed method effectively identifies distribution network topology and line parameters，exhibits strong robustness，and adapts well to various complex operating conditions.
Fault line selection method for high-resistance ground faults in asymmetric distribution networks with ne utral point voltage source connection
Tao Tang, Silin Xiang, Qilong Rong, and Kangjian Yuan
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.005
A fault line selection method based on neutral point voltage source connection for single-phase high-resistance ground faults is proposed to address the challenge of single-phase high-resistance ground fault identification in asymmetric distribution networks.Firstly，the zero-sequence current component characteristics of lines in asymmetric distribution networks are analyzed，and an equivalent model for the zero-sequence network of asymmetric distribution networks is built.Secondly，when the zero-sequence voltage exceeds its limit，a high-frequency voltage source is connected to the system neutral point，with the influence of asymmetric parameters of each line eliminated by calculating the asymmetric current.The ratio of the power frequency component to the high-frequency component of the line zero-sequence current is denoted as the power-to-high ratio （PHR ）.It is found that the PHR of the fault line is significantly higher than that of the healthy line；the line with the largest PHR is identified as the fault line；the high-resistance fault line selection can be realized.At the same time，the influence of transition resistance and the neutral point connection on the line selection criterion is analyzed，with simulation analysis carried out based on case studies.The results demonstrate that the proposed method effectively eliminates the influence of three-phase distribution parameter asymmetry in high-resistance fault line selection，and exhibits robust resilience against transition resistance.Moreover，it is not affected by the neutral point connection and has certain applicability to arc high-resistance grounding.
Cause analysis of low voltage in distribution networks based on multi-scale time-frequency features and global dependency modeling
Peng Chen, Sihan Zhou, Liping Fan, and Xiaosheng Yu
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.006
A cause analysis model based on multi-scale time-frequency features and global dependency modeling is proposed to improve the accuracy and robustness of cause classification and address the complex and diverse causes of low voltage in distribution networks and the non-stationary nature of signal characteristics.The model consists of two feature extraction branches：the wavelet convolution branch employs one-dimensional discrete wavelet convolution to extract multi-scale time-frequency features and capture the local dynamic variations of the signals；the Transformer branch converts the time-series data into a recursive graph and models global temporal dependency through the self-attention mechanism.Subsequently，local and global features are integrated through feature fusion and input into a multi-label classifier to complete the cause classification task.Finally，on the low-voltage dataset of a distribution dataset from a certain city in China，simulation analysis of the model is carried out.Simulation analysis shows that the model achieves a classification accuracy of 98.74%.Compared to traditional methods，this model effectively combines local and global features and significantly improves cause classification capability，demonstrating excellent performance and robustness in handling non-stationary signals.The model can provide a reliable solution for the accurate analysis of low voltage problems in distribution networks.
Calculation method for distribution network line loss based on feature selection and LSTM-XGBoost-SHAP
Li Yang, Guo Wang, and Yongzhi Min
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.007
Accurately calculating distribution network line losses is crucial for ensuring the economic，safe and stable operation of the power system.The conventional calculation methods cannot effectively utilize electrical feature parameters and involve cumbersome calculation procedures.To address this issue，the study proposes an interpretable calculation method for distribution network line loss based on electrical feature selection and a hybrid model integrating long short-term memory （LSTM ） and extreme gradient boosting-SHapley Additive exPlanations （XGBoost-SHAP ）.In this method，theoretical electrical feature indicators for line loss are determined by comprehensively considering the relationships between various electrical parameters and distribution network line losses.A feature selection method based on cross-correlation is proposed to eliminate redundant features and construct the optimal feature subset，thereby overcoming high redundancy among features of the same type.Then，an LSTM-XGBoost-SHAP interpretable calculation framework is developed to solve the problem of insufficient utilization of heterogeneous data by conventional single models.In this framework，the de ep features of time series data are extracted using the LSTM network，and the time series features are fused with static features by the XGBoost model to achieve high-precision calculation of the theoretical line loss.The contribution of each input electrical feature parameter to the theoretical line loss is analyzed using the SHAP module.Finally，the accuracy of the proposed method is verified using actual operational data from a distribution network in Northwest China.The results show that the proposed method achieves a root mean square error （RMSE ） of less than 0.213% and a mean absolute percentage error （MAPE ） of not more than 5.634%，indicating significantly better calculation accuracy than that of the conventional single models as well as favorable robustness and engineering applicability.
Multi-entity multi-measure combined line-loss control for active distribution networks
Haiyun An, Yufang Liu, Qian Zhou, and Jiaxun Li
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.008
The large-scale integration of diverse entities，including wind power，photovoltaic generation，and energy storage，has transformed distribution networks from traditional passive networks into active bidirectional networks.While helping alleviate energy and environmental crises，this transformation also poses significant challenges to line-loss control in active distribution networks （ADNs ）.To address this issue，this paper proposes a multi-measure combined line-loss control method considering the participation of multiple entities.First，various loss-reduction measures，including distribution line upgrading，transformer replacement，and reactive power compensation，are considered comprehensively.A multi-objective optimization model accounting for both loss-reduction effectiveness and economic performance is established，with optimal economic operation of the ADN as the objective and secure operation of the distribution network with multi-entity participation as the constraint.Second，a multi-measure combined line-loss control strategy based on the benefit coefficient is proposed.By quantifying the ratio of loss-reduction benefits to costs，the optimal combination of integrated loss-reduction measures is determined.Third，a benefit-coefficient-based solution method is proposed to solve the established optimization model and obtain the optimal integrated loss-reduction strategy that meets the predefined loss-reduction target of the distribution network.Finally，simulation and comparative analyses were conducted through a case study.The simulation results show that the proposed method significantly reduces the line-loss rate of the multi-entity distribution network and maximizes the loss-reduction benefits.This study provides a scientific and efficient basis for decision-making on loss-reduction retrofits in ADNs.
Short-term bus load forecasting considering interval forecasting information and temporal convolutional network
Xinming Liu, Haichen Li, Jianguang Ji, Shuo Sun, and Junxin Bian
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.009
The bus load forecasting in substations is essential for the stable operation of power systems.With the large-scale integration of renewable energy generation into the grid，low-voltage bus loads become increasingly volatile，and their uncertainty can hardly be quantified by point forecasting alone.Interval forecasting can enhance decision-making in power dispatching.For this reason，the study proposes a hybrid short-term bus load forecasting method that integrates point and interval forecasting.Specifically，complementary ensemble empirical mode decomposition with adaptive noise （CEEMDAN ） is applied for primary decomposition of the original bus load.The features of the highly volatile high-frequency components from primary decomposition are extracted through secondary decomposition of the high-frequency components using variational mode decomposition （VMD ）.Next，the sample entropies （SEs） of the subcomponents are calculated to assess their complexity and reconstruct them into three groups of representative components.These representative components are then input into an improved sparrow search algorithm and temporal convolutional network （ISSA-TCN ） model，which incorporates Tent chaotic mapping，adaptive weights，Levy flights，and variable spiral search strategies，to enhance point forecasting accuracy.Finally，quantile transformation （QT） is applied to smooth the abnormally distributed errors in kernel density estimation （KDE ），and confidence interval forecasting is performed to obtain probabilistic intervals.Comparative experiments with real bus load data show that compared with both single and hybrid models，the proposed point forecasting method reduces the mean absolute error to 3.52% and improves the fitting degree to 94.62%.Compared with parametric and nonparametric methods，the proposed interval forecasting method lowers the mean interval score （MIS） to 4.2 at 95% confidence，improves the prediction interval coverage probability （PICP ） to 98%，and covers the largest number of actual values.The proposed method effectively quantifies the uncertainty of point forecasting，comprehensively reflects the variation patterns of bus loads，and facilitates the operation and management of power systems.
Demand response capability evaluation of central air conditioning based on random forest
Tingzhe Pan, Jinhua He, Zijie Meng, Xin Jin, Wei Zhou, Wangzhang Cao, and Chao Li
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.010
Central air conditioning in commercial buildings is a high-quality controllable resource，and the rational utilization of its regulation capacity is of significance to the supply-demand balance of power systems.In particular，during summer peak electricity consumption periods，effective regulation of massive central air conditioning loads can greatly relieve the pressure of source-load balance in power grids.To this end，a demand response capability evaluation of central air conditioning based on random forest （RF） is proposed.Assuming that the participation of central air conditioning in demand response is limited to controlling the on/off status of the central refrigeration system，then the demand response capability can be characterized by the maximum on/off duration starting from any given moment.Firstly，the chilled water circulation process，indoor thermal dynamics，and on/off control logic of central refrigeration systems for central air conditioning are sorted out，and the key state variables affecting the maximum on/off duration are clarified.Secondly，a dataset containing indoor and outdoor temperatures，historical supply and return temperatures of chilled water，and maximum on/off duration is established based on physical model simulation or field measurement data.Then，an RF model is trained with the above state variables as inputs and the maximum on/off duration as outputs.Finally，case comparison and comfort verification are conducted to validate the evaluation accuracy and applicability of the proposed method.The proposed RF model avoids reliance on the complex thermal dynamics parameters of buildings and features low implementation barriers.Simulation comparison and analysis demonstrate that the proposed method exhibits high accuracy in evaluating the maximum on/off duration of central air conditioning，and it will not significantly reduce indoor occupants ’ comfort when assessing demand response capacity.
Identification of electric bicycle charging load based on label-guided attention mechanism
Xianyi Cui, Weijie Zeng, Mouhai Liu, Liman Shen, Yeqin Ma, Yuge Zhang, and Yunpeng Gao
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.011
The growing popularity of electric bicycles is accompanied by the risk of charging-related fires，calling for prompt effective monitoring.Non-intrusive load monitoring （NILM ） provides a key technical approach for this purpose.A deep learning identification algorithm based on label-guided attention is proposed for non-intrusive identification of electric bicycle charging loads.Specifically，a feature sample library is constructed by collecting high-frequency power data from electric bicycles and various household appliances.Then，deep features are extracted using a one-dimensional convolutional neural network （1DCNN ），and feature representation is enhanced by combining feature and position attention mechanisms.Furthermore，sequential dependencies are captured using a bidirectional recurrent neural network （BRNN ） to achieve accurate load forecasting.Finally，simulation analysis is conducted using the publicly available plug-load appliance identification dataset （PLAID ） and actual data of electric bicycle charging loads to verify the effectiveness of the proposed method.The results show that the proposed method performs excellently in the identification task，delivering results superior to those of the existing models.
Lightweight method for remaining useful life prediction of instrument transformers based on quantization-aware training
Wei Zhang, Jiaona Mao, Zhuolin Bao, Zhuo Long, and Xiaofei Zhang
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.012
Instrument transformers are key measurement devices in power systems，and their performance directly affects the safe and stable operation of power grids.As deep learning is widely applied to device state prediction，the high computational costs and storage requirements brought by expanding model scales severely restrict their deployment on edge devices.To this end，an adaptive quantization-aware training （QAT ） method，named minimizing quantization error-aware training （MinQuant ），is proposed.Firstly，an adaptive quantizer that automatically adjusts quantization parameters according to data distribution characteristics is designed.Secondly，a quantization error minimization algorithm is proposed to explicitly optimize quantization error during training and achieve collaborative optimization of model performance and quantization precision.Finally，the proposed method is validated on a voltage transformer secondary voltage dataset provided by a power grid company.The results show that the proposed method achieves almost no precision loss under 8-bit quantization and maintains high predic tion precision compared to existing methods in low-bit quantization scenarios such as 4-bit，providing an effective solution for lightweight instrument transformer prediction models.
Exploration of Lunar Surface Energy System
Review of lunar surface energy system technologies
Ming Zhang, Li Lu, Yaodong Liu, Hong Du, Chengxiong Tang, Hua Geng, Qiaoling Tong, and Wei Lu
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.013
According to China ’s lunar exploration plan，China aims to achieve a manned lunar landing before 2030 and establish the basic model of a lunar research station before 2035.As essential infrastructure for the construction and operation of the lunar research station，the energy system is fundamental to ensuring its long-term，stable operation.This study surveys domestic and international lunar surface energy systems and analyzes the energy system requirements of China ’s lunar research station.Based on this analysis，the application scenarios and development requirements for power generation，energy storage，power management，and power transmission and distribution technologies are discussed in detail.Various types of energy systems are proposed and compared.On this basis，an architecture for the lunar surface energy system and its key technologies are proposed to provide a reference for the future development of lunar surface energy systems.
Design and analysis of power supply transmission for lunar surface energy system
Ming Zhang, g Jiankun Zhan, Yuxiang Yang, Chengxiong Tang, Lijun Ma, Ji ’nan Ma, Hong Du, Hua Geng, and Yu Zhao
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.014
The energy system is an important component of the lunar research station ’s infrastructure and is responsible for providing long-term，stable electric power.Based on the construction stages and mission requirements of the lunar research station，this paper briefly introduces the composition of the energy system，proposes an energy transmission topology architecture based on a distributed system，and establishes mathematical models for the power conversion and transmission links.Considering the characteristics of AC and DC systems，as well as factors such as system weight，reliability，and the application of space high-voltage technologies，this paper compares and analyzes the characteristics of transmission buses under different systems，voltage levels，and transmission power levels，determines the optimal transmission voltage，and presents the corresponding technical analysis.The results of this study can provide references for the design of energy transmission topology architectures and transmission links for extraterrestrial bases such as lunar research stations.
Famous Host
Economic-security trade-off methods for optimal dispatch in new power systems : review and prospects
Lei Yan, Guocan Yan, Chuangxin Guo, Yilin Zhou, Yan Meng, Rouyu Lin, Keyi Chen, and Zuyi Li
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.015
The strong uncertainty of new power systems makes the economic-security trade-off in day-ahead optimal dispatch more challenging.From the perspective of the interaction between risk preferences and optimization models，this paper categorizes trade-off methods into three modes：the a priori setting mode，the a posteriori selection mode，and the endogenous trade-off mode.First，it analyzes the limitations of the a priori setting mode，which relies on subjective experience to determine risk parameters，such as risk levels and uncertainty budgets，as well as the computational challenges of the a posteriori selection mode in high-dimensional risk settings.Second，it focuses on the endogenous trade-off mode represented by adjustable chance constraints，explaining how this mode endogenizes risk parameters as decision variables and introduces risk cost functions to achieve an adaptive optimal economic-security trade-off.Furthermore，a simplified numerical example illustrates the potential advantages of the endogenous trade-off mode，followed by a multidimensional comparison of the three modes.Finally，future research directions are outlined with respect to key challenges in risk characterization，model transformation，and efficient solution methods.
Transient rotor angle stability impact assessment method based on dominant mode identification for high-penetration renewable energy power systems
Jinlong Zhang, Mengjie Yu, Yanhong Bao, Yuxiang Zhao, Jiaqi Song, Yingping Shi, and Zeheng Sun
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.016
This paper addresses the new challenges in the quantitative assessment of transient rotor angle stability in high-penetration renewable energy power systems，proposing a method for evaluating the impact degree based on dominant mode identification.First，the limitations of traditional “impact degree ” assessment methods based on a single-instant kinetic energy approximation are discussed in terms of their ability to address system diversification，low inertia，and differences in the dynamic mechanisms of the two stages.It is also demonstrated that accurate generator grouping is an important prerequisite for effective assessment.On this basis，a transient energy function reflecting the relative motion between generator groups is constructed，and a theoretical connection between impact degree and transient energy is established.Subsequently，global statistical characteristics are extracted from the generators ’ relative rotor angle trajectories to accurately identify the dominant disturbance mode and reliably group the generators.Finally，on the basis of accurate grouping and by comprehensively considering the contributions of accelerating energy during the fault and decelerating energy after the fault at two key characteristic moments，a new comprehensive impact degree assessment index is constructed.Analysis results obtained from a regional power grid simulation system demonstrate that the proposed method can effectively identify the dominant oscillation mode and accurately assess the impact of different generators on the system ’s transient stability，providing a new quantitative tool for the security and stability analysis and precise control of high-penetration renewable energy power systems.
Unit commitment optimization method based on linear approximation of AC power flow
Shengbo Ji, Jinlong Zhang, Ruipeng Guo, and Chuangxin Guo
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.017
To address the conflict between the neglect of reactive power and voltage issues in DC power flow and the complexity of solving accurate AC power flow in traditional transmission network unit commitment，this paper proposes a coordinated active and reactive power optimization strategy based on a linear approximation of AC power flow.First，the AC power flow equations are linearized，transforming the nonlinear unit commitment problem into a Mixed-Integer Quadratic Programming （MIQP ） model.The model is further transformed into a mixed-integer linear programming （MILP ） model through piecewise linearization of the cost function.Second，considering that reactive power and voltage relationships in power systems are not approximately linear over a wide range，a base point-increment method is introduced to improve the accuracy of the linearized model.Finally，case studies are conducted on the IEEE 39-bus and IEEE 118-bus systems.GUROBI is employed to solve the model efficiently，and accurate power flow results obtained using MATPOWER are used to further verify the effectiveness of the proposed method.The results demonstrate that the proposed method achieves coordinated active and reactive power optimization in systems of different scales.The application of the base point-increment method and piecewise linearization of the cost function significantly improves solution performance and provides reliable and efficient unit commitment schedules.
Data-driven distributionally robust scheduling method for park-level integrated energy system considering electric vehicle integration
Chuangxin Guo, Kun Chen, Yumian Lin, Houbo Xiong, Lei Yan, and Liwei Du
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.018
Fluctuations in renewable energy output and the growing and stochastic electric vehicle （EV） charging demand introduce source- and load-side uncertainties that significantly affect the scheduling plan of the park-level integrated energy system （PIES ）.To improve the economy and robustness of PIES operation，this paper proposes a data-driven distributionally robust scheduling method for PIES considering EV integration.First，given that the uncertainty associated with EV integration is difficult to characterize accurately，five features of EV charging behavior are selected.A Gaussian mixture model is used to characterize the multimodal behavior of EV charging，and an uncertainty set is constructed by combining Neyman sampling with comprehensive norm constraints.Second，a coordinated delayed and staggered charging management scheme is designed to achieve peak shaving and valley filling while meeting users ’ charging requirements.Then，EV charging management is incorporated into the economic scheduling framework of PIES to construct a data-driven distributionally robust optimization （DRO ） model.Finally，an improved column-and-constraint generation algorithm is proposed to solve the DRO scheduling problem.The case study results show that the proposed delayed charging management scheme smooths the PIES electric load curve，significantly increases the renewable energy accommodation rate，and reduces dependence on purchased electricity.Meanwhile，compared with a conventional two-stage robust optimization model，the proposed method reduces the total cost by 10.26%，verifying its feasibility and economic efficiency and providing a new pathway for efficient PIES scheduling under high EV penetration.
Clean energy and energy storage
Hybrid HMM-MGAN approach for long -term renewable energy output scenario generation
Fang Liu, Yang Liu, Ruichen Hao, Xiang Wang, Yuyan Song, Yunche Su, and Haibo Li
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.019
To address the limitations of existing numeric al models for long-term renewable energy scenario generation，such as oversimplified structures，poor output quality，a predominant focus on seasonal trends，and insufficient modeling of output details，this paper proposes a long-term renewable power scenario generation method based on a dual-layer weighted ensemble strategy.The proposed approach integrates multiple power reconstruction models to achieve high-fidelity reconstruction of historical renewable energy output.On this basis，a hidden Markov model （HMM ） is employed on a monthly basis to construct daily-scale state transition sequences of renewable energy output，while a modified generative adversarial network （MGAN ） is used to generate monthly renewable energy output curves.The monthly output curves are then concatenated month by month to generate annual renewable energy output scenarios covering 8 760 h.To further enhance the quality of the generated scenarios，the MGAN model is augmented with a volatility constraint and a self-attention mechanism.In addition，an information redundancy mechanism is designed to mitigate boundary effects introduced by convolutional operations.Finally，case studies are conducted using actual historical wind and solar power output data and multi-year meteorological data from a province in southwestern China.The results show that the proposed method offers significant advantages in terms of scenario generation accuracy and the ability to reconstruct output details.
Coordinated suppression method for power system inter -area oscillations based on supplementary damping control of DFIGs
Jun Xiao, You Situ, Anping Huang, and Runfeng Hu
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.020
Doubly fed induction generator （DFIG ）-based wind turbines have been proven to be effective devices for damping inter-area low-frequency oscillations due to their flexible power regulation capability.However，due to the difficulty of accurately modeling power systems，the complexity and variability of operating conditions，the existence of substantial uncertainties and disturbances，and the coupling of multiple oscillatory modes，it is difficult for traditional model-based supplementary damping control methods with fixed parameters for DFIGs to achieve good damping performance.Therefore，this paper proposes a data-driven coordinated supplementary damping control method for DFIGs.First，partial-form dynamic linearization is used to establish a decoupled multi-input multi-output data-driven model that characterizes the dynamic behavior of a power system with multiple oscillatory modes.Meanwhile，the control parameters are updated in real time according to the I/O data of the power system to adapt to the changing operating conditions of the system.A discrete sliding-mode surface function is then designed to generate a control law that damps multiple oscillations while effectively suppressing the effects of uncertainties and disturbances，as well as coupling among multiple oscillatory modes，on damping performance.Finally，the stability of the system under the proposed method is analyzed，and comparative simulations demonstrate that the proposed method effectively suppresses multimode oscillations while exhibiting strong adaptability and robustness.
Relocation and charging/discharging power optimization of modular energy storage for voltage limit violation mitigation in rural photovoltaic distribution station areas
Haifeng Yu, Yifei Wang, Yu Li, Hanhang Yin, Siyuan Peng, Puling Huang, and Zheng Peng
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.021
To address the seasonal overvoltage and undervoltage issues in China ’s rural photovoltaic （PV） distribution station areas，this study proposes a relocation and charging/discharging power optimization method for modular energy storage adapted to the mitigation of voltage limit violations in the rural PV distribution station areas.Voltage limit violations are mitigated by relocating modular energy storage units to PV distribution station areas and regulating the active charging/discharging power of the energy storage units and the reactive power of the inverters.Specifically，the source-load characteristics of China's rural PV distribution station areas are analyzed.The feeder voltage equation is applied to reveal that the voltage limit violations in the rural distribution station areas are caused by the seasonal mismatch between sources and loads.Next，a mathematical model of modular energy storage is developed by leveraging its detachable and flexible deployment capabilities.Then，an optimization model for the relocation and charging/discharging power control of modular energy storage considering the relocation and leasing costs of modular energy storage is constructed on the premise of reducing node voltage deviations in the distribution network，network losses，and PV curtailment penalty.The model optimizes the candidate nodes for energy storage relocation，the number of relocated units，and their active and reactive power output.Finally，the feasibility and effectiveness of the proposed method are verified on the power distribution system in Hengdong County，Hunan Province.The results show that modular energy storage can effectively improve the voltage quality and economic benefits of China's rural PV distribution station areas.
Accurate calculation method for transient overvoltage in multi-infeed renewable power systems under asymmetric faults
Lu Zhang, Jie rui Huang, Bo Tang, and Chenxi Yang
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.022
During fault recovery，transient overvoltages （TOVs ） have become an increasingly prominent issue in power systems with high penetration of renewable energy sources.However，related research，primarily focusing on three-phase short-circuit faults，rarely addresses asymmetric fault scenarios.This study approaches the issue by analyzing the transient characteristics of renewable energy units under asymmetric faults.Then，it presents an equivalent model incorporating negative-sequence control and compares the severity of TOVs caused by three-phase symmetric and asymmetric short-circuits in a single-unit system.Furthermore，to address the difficulty in characterizing the coupling between generation units and between the units and negative-sequence control in multi-infeed systems，the study proposes an iterative TOV calculation method based on positive- and negative-sequence components.This method models renewable energy units with negative-sequence control as voltage-controlled current sources and accurately evaluates the overvoltages during the fault and after fault clearance by iteratively solving port voltages and injected currents.Finally，electromagnetic transient simulations are conducted on the PSCAD/EMTDC numerical simulation and analysis platform to verify the effectiveness of the proposed method.This method can serve as a reference for calculating TOVs in multi-infeed renewable power systems under asymmetric faults.
Distribution network energy storage siting planning based on predictive load generation using Prophet and GAN
Jiayu Xu, Xiaodong Chen, Yafeng Wang, Zhaoyan Liu, Yiyu Gong, Fanglan Liu, and Jiaoxin Jia
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.023
An energy storage siting and sizing planning method based on load trend extraction and forecasting is proposed to address the challenges of photovoltaic （PV） curtailment and bidirectional power flow overloads posed by distribution networks with high PV penetration.Firstly，the Prophet algorithm is adopted to decompose the net load of each node into trends and seasonal components，and the residual load is obtained.Secondly，a generative adversarial network （GAN ） is employed to model the relatively random residual component and generate load forecast data for the following year.The generated residual data are then combined with the trend and seasonal components to obtain a new annual net load that conforms to the original statistical patterns.Thirdly，based on the forecasted load data，an improved Bayesian optimization algorithm and nonlinear programming are adopted for energy storage siting and sizing to obtain the optimal energy storage configuration scheme.Finally，simulations are conducted based on real data from the following year to verify the predictability and economic efficiency of the proposed model.This study can provide reference for energy storage siting planning of distribution networks.
Optimal regional allocation of mobile energy storage in urban distribution networks considering distribution-transportation coupling
Zifa Liu, Chang Wang, and Mengni Ye
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.024
Given the increasing demand for flexible resources in distribution networks，investigating the trade-off between the economic efficiency of resource allocation during the pre-planning phase and the enhancement of resilience in the fault recovery stage is essential for avoiding asset underutilization and excessive investment.To address this issue，the paper proposes a regional allocation method for mobile energy storage systems （MESSs ） that considers power-transportation coupling.First，a coupled distribution-transportation model is constructed based on the interactions among vehicles，charging stations，and the distribution network to characterize traffic flow dynamics.Next，a comprehensive indicator integrating electrical modularity and geographic accessibility is developed and an improved genetic algorithm is employed to partition the coupled distribution-transportation network into clusters.A two-layer optimization framework integrating regional planning and scheduling operation is then designed，in which the upper-layer model optimizes MESS capacity and power configurations while the lower-layer model formulates dynamic scheduling strategies based on spatiotemporal mobility characteristics of the MESSs.Particle swarm optimization （PSO） and mixed-integer linear programming （MILP ） are further employed to solve the planning model and obtain the regional allocation scheme for the MESSs.Finally，simulation analysis of the proposed method is conducted through case studies.The results demonstrate that the proposed method not only effectively reduces the complexity of spatiotemporal MESS scheduling but also significantly improves the operational economy of the distribution network.It effectively promotes the coordination between economic efficiency optimization and resilience enhancement during the pre-planning phase.
Robust bidding strategy analysis for charging stations considering response uncertainty
Xiaofeng Jiang, Bo Zhou, Chun Tang, Yunyang Xu, Wei Wei, Xinwei Sun, and Jianwei Yang
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.025
The large-scale integration and interaction of electric vehicles with the power grid have become an inevitable trend.As key entities in the interaction between electric vehicles and the power grid，charging stations can aggregate the charging and discharging resources of electric vehicles to participate in electricity market bidding，thus balancing their own interests and the operational needs of the power grid.Based on this，a model capable of predicting the day-ahead responsive capacity was first proposed by considering the active charging response of electric vehicles.Second，a robust optimization bidding model was established for charging stations participating in the day-ahead electricity market，considering the effects of uncertainties in EV charging and discharging behavior on their bidding strategies and treating the maximum charging and discharging powers as uncertain variables.Finally，simulations were conducted to obtain the charging station ’s day-ahead bidding scheme based on the proposed model，and the effect of the model on the day-ahead bidding strategy was verified in the real-time market.The simulation results demonstrate that the proposed strategy can effectively enhance robustness while ensuring the economic efficiency of the charging station ’s day-ahead bidding.In addition，the strategy yields a real-time market bidding plan that conforms to the actual charging and discharging behavior of the charging station.This strategy can provide a reference for the robust bidding of charging stations.
High voltage and insulation
Inversion model for point of maximum vibration in oil-immersed transformer based on multiphysics load-transfer characteristics
Xingxing Hao, Haiying Li, and Jiancheng Song
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.026
Accurately obtaining data at the point of maximum vibration in a transformer core is one of the challenges in power equipment health management.To address this problem，a new method based on multiphysics modeling and inversion techniques was proposed for detecting the point of maximum vibration acceleration in a transformer.First，a multiphysics load-transfer coupling model incorporating electromagnetic，structural，and fluid fields was used to analyze the vibration mechanism under an alternating and nonuniformly distributed magnetic field.The point of maximum i nternal vibration was then determined through harmonic response analysis.Second，considering data correlations，vibration measurement points on the outer wall of the oil tank that were strongly correlated with the point of maximum vibration in the core were selected.A random forest algorithm optimized through K-fold cross-validation and grid search was used to establish an inversion framework for the point of maximum vibration，enabling accurate and timely detection of the internal vibration state.Finally，the proposed method was validated using a simulation model of a 35 kV three-phase oil-immersed transformer.The results show that the multiphysics simulation samples generated under various operating conditions are valid and reliable.The root mean square error （RMSE ） of the inversion model for the point of maximum vibration is 1.2%，and the coefficient of determination is 0.92，thus accurately reflecting the internal vibration state.
SF6 gas leakage detection method for GIS equipment based on improved YOLO 11 model
Cheng Zeng, Shaosheng Fan, Jiashun Xie, and Kaiyu Guo
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.027
An SF6 gas leakage detection method for GIS equipment based on an improved YOLO 11 model is proposed to address the problem of internal SF6 gas leakage in outdoor substation gas insulated switchgear （GIS） equipment caused by environmental factors and equipment aging.The Real-ESRGAN super-resolution generative adversarial network is adopted for denoising and enhancement，and then the YOLO 11 model is improved to solve the problems of missed detection of small gas plumes，false detection due to cloud interference，and inability to perform multi-task detection in the original YOLO 11 model.The improved model increases the number of detection heads to four and incorporates a GAM attention mechanism based on the YOLO 11 model，replacing the original detection heads with fused detection heads to simultaneously accomplish segmentation detection and keypoint detection tasks.Finally，simulation experiments as well as simulation and field experiments carried out by a self-developed SF6 gas leakage inspection robot are performed to verify the feasibility and effectiveness of the proposed method.Simulation experiments show that the improved model achieves ImAP 50 and ImAP 50-95 values of 85.2% and 54.4% for the segmentation task，and 96.3% and 96.5% for the keypoint task，respectively，outperforming the YOLO 11 model.Simulation and field experiments using a self-developed SF6 gas leakage inspection robot further verify the leakage detection effectiveness of the improved model，demonstrating that the proposed method can replace manual labor in gas leakage detection.
Multimodal feature fusion method for identifying excitation inrush current in distribution transformer
Fan Yan, Luxin Zhan, Chun Chen, Yi An, and Yijia Cao
Date posted: 9-1-2026
DOI: https://doi.org/10.19781/j.issn.1673-9140.2026.04.028
With the widespread integration of high-penetration distributed generation （DG） and power electronic devices，the harmonic components of fault currents have become increasingly prominent，preventing traditional relay protection methods from effectively distinguishing between excitation inrush currents and fault currents.Moreover，existing time-frequency analysis methods rely on manually selected evaluation metrics，which often lead to misjudgments.To address these issues，this paper proposes a deep learning-based identification method that fuses image- and statistical-domain features，in which the neural network adaptively learns features for identification.Specifically，the Gramian angular field （GAF ） method is employed to transform time-series data into images，from which image features are extracted using ResNet 18.Then，statistical-domain features of the fault current are extracted，and the two types of features are fused through low-rank multimodal fusion （LMF ）.Furthermore，a multilayer perceptron （MLP ） is used to train the fused multimodal features，thereby efficiently identify fault currents and excitation inrush currents.Finally，simulation experiments are conducted to verify the feasibility and effectiveness of the proposed method.The results indicate that the proposed method achieves an identification accuracy of up to 99.72% and can be further applied to classify fault currents.
