30 June 2026, Volume 46 Issue 3

  

• Select all
  |

• Select
  Research on the Optimal Allocation of Hybrid Energy Storage Capacity Considering Deadband

  JIN Enshu, WANG Jiachen, YANG Dejian, ZHENG Taiying

  2026, 46(3): 1-8. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0001-09

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  To cope with the uncertainty brought by wind power integration,this paper proposes a hybrid energy storage optimization configuration strategy that combines supercapacitors and lithium batteries for coordinated stabilization,in order to improve the stabilization effect of wind power fluctuations and take into account the impact of stabilization dead zones.Based on typical daily wind power data,ensemble empirical mode decomposition （EEMD）is applied to decompose wind power signals into direct grid connected components and energy storage smoothing components；Secondly,secondary EEMD is used to decompose the energy storage power into the high-frequency component smoothed by supercapacitors and the low-frequency component smoothed by lithium batteries.A control strategy that takes into account the smoothing dead zone is proposed to address the impact of frequent charging and discharging on energy storage life,with the aim of extending equipment lifespan.This article constructs three capacity optimization configuration models that take into account the smoothing dead zone,and obtains the comprehensive cost and state of charge of the energy storage system by solving them in different scenarios.Evaluate the depth of charge and discharge using rainflow counting method and calculate the equivalent cycle life.Finally,the effectiveness of the strategy was verified by comparing the number of charge and discharge cycles and energy storage life.
• Select
  An Improved SSD-Based Method for Insulator Defect Detection 

  WANG Liang, CAI Dongxue, BAO Meiling, Wu Lingiao, WEI Enyu, ZHANG Hui

  2026, 46(3): 9-19. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0009-11

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  To address the problems of complex backgrounds,small defect regions,susceptibility to adverse weather conditions,and the resulting low image clarity and loss of detail in transmission line insulator images,an improved SSD Single Shot MultiBox Detector）network is designed for efficient detection of insulator defects.By performing detection on feature maps at different levels,SSD can capture multi-scale targets and is more suitable for detecting small objects and subtle defects.An improved generative adversarial network is adopted to enhance image quality under rainy, snowy,foggy,and nighttime conditions.MobileNetV3 is used to replace VGG16 to achieve model light weighting and improve detection speed.In addition,CARFB （Coordinate Attention and Receptive Field Block）and SE （Squeeze-and-Excitation）attention modules are introduced into the feature layers to enhance the recognition ability for small defects such as cracks and contamination.Experimental results show that the proposed method achieves a detection precision of 93.4%and an mAP mean Average Precision）of 95.1%，enabling efficient detection of insulator defects while maintaining high mAP and recall.
• Select
  Optimal Scheduling Model of Production Process of Short-Process Steel Enterprises Based on Source-Load Synergy

  LI Junhui, ZHANG Jiong, LIU Hengyu, YU Meng, LI Yan, HU Dawei, ZHU Xingxu, LI Cuiping

  2026, 46(3): 20-29. https://doi.org/10.19718/i.issn.1005-2992.2026-03-0020-10

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  In view of the problem that the complex process requirements and strict safety constraints in the production process of short-process steel enterprises lead to the limited effect of source-load synergy in industrial parks,a sourceload synergy optimization scheduling model considering the production process of short-process steel enterprises was proposed.Firstly,the industrial park architecture integrating the production equipment batches of short-process steel enterprises was designed,and the task-process-warehouse coupling model of the production process of short-process steel enterprises was established based on the state task network,so as to explore the potential of production batch adjustment and optimization of short-process steel enterprises under the background of source-load synergy.Secondly,a time-of-time peak electricity price joint response mechanism is proposed,and a source-load collaborative optimization scheduling model with the goal of minimizing the operating cost of the industrial park is constructed,and the output constraints of generator sets and the production process constraints of short-process steel enterprises are considered in the process.Finally, based on the production data of the industrial park,the results show that the proposed model can optimize the batch of production equipment of short-process steel enterprises under the premise of ensuring production safety,so as to improve the operation economy of the industrial park by 14.96%and the utilization rate of new energy by 4.56%.
• Select
  Optimization Scheduling of MECS Considering Carbon-Green Certificate Dual-Tier Mutual Recognition and Generalized Energy Storage 

  SU Jiantao, GENG Kailun, CAI Tingting

  2026, 46(3): 30-42. https://doi.org/10.19718/i.issn.1005-2992.2026-03-0030-13

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  To enhance the flexibility and low-carbon economic performance of multi-energy coupled systems （MECS）under high penetration of renewable energy,this study proposes an optimal scheduling strategy for electric-thermal-gas-hydrogen MECS,considering carbon-green certificate dual-tier recognition and generalized energy storage.Firstly,the utilization pathways of hydrogen are analyzed,and a refined multi-utilization MECS model incorporating hydrogen is established.Secondly,a generalized energy storage framework is introduced,which integrates electrical,thermal,and hydrogen storage with flexible loads and flexibly operated carbon capture power plants,enabling the coordinated optimization of multiple types of storage resources and enhancing the system's inter-temporal regulation capability.Finally,a carbon-green certificate dual-tier recognition mechanism is constructed by combining tiered green certificate trading with tiered carbon trading,and a low-carbon economic dispatch model is formulated to minimize the total cost, including green certificate trading cost,carbon trading cost,energy procurement cost,wind curtailment cost,and virtual storage cost.Case study results indicate that the proposed model reduces the total system cost of the MECS by 39.2%， decreases carbon emissions by 63.7%，and achieves full wind power utilization,demonstrating that considering carbongreen certificate dual-tier recognition and generalized energy storage can effectively improve the economic and lowcarbon performance of multi-energy coupled systems,and provides a feasible pathway for their low-carbon and clean transition.
• Select
  Transformer Fault Diagnosis Model Based on Improved Sparrow Search Algorithm Optimizing Convolutional Neural Networks

  SHANG Haikun, LIU Zhidong, WANG Jinjuan

  2026, 46(3): 43-52. https://doi.org/10.19718/i.issn.1005-2992.2026-03-0043-10

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  This paper proposes an Improved Sparrow Search Algorithm （ISSA）to optimize the Convolutional Neural Networks （CNN）for transformer fault diagnosis,addressing the issues of low diagnostic rate and difficult parameter selection in traditional transformer fault diagnosis.Firstly,the paper utilizes an optimization strategy to obtain the ISSA and applies it to optimize the parameters of CNN.Then,based on the dissolved gas content data in the original transformer oil,high-dimensional gas ratio features are calculated.Subsequently,a combination of contrast entropy and low variance filtering is used for dimensionality reduction.Finally,using the selected low-dimensional data features,the ISSA-CNN transformer fault diagnosis model is established.The fault diagnosis experiments of ISSA-CNN,SSA-CNN,and PSO-CNN are conducted using a 5-fold cross-validation method.The results show that the proposed model has better diagnostic performance and generalization ability.
• Select
  Planning and Carrying Capacity Enhancement Method of PV-Storage Charging Stations Integrating Federated Learning and Zonotope

  SUN Liang, LU Yanpeng, Wu Jinrong, LI Zhuojun, LIU Jiaao

  2026, 46(3): 53-61. https://doi.org/10.19718/i.issn.1005-2992.2026-03-0053-09

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  To address the challenges posed by the high penetration of electric vehicles （EV）to distribution network planning,this paper proposes a bi-level planning method for photovoltaic-storage （PV-ESS）charging stations and distribution networks that considers the spatiotemporal characteristics of charging loads.Firstly,to protect user data privacy and improve prediction accuracy,an improved hybrid prediction framework combining federated graph neural networks and Transformer （FSTG-T）is constructed,and an improved Gaussian mixture model （GMM）is used to accurately quantify the randomness of users in charging time and power selection.To further enhance the carrying capacity of distribution networks,a three-dimensional evaluation index system is established,including improved stability （Power Voltage Stability Margin,PVSM），flexibility （Demand Response Capacity Margin,DRCM），and economy （Marginal Comprehensive Benefit Index for Carrying Capacity Enhancement,MCBI-CCE），which comprehensively quantifies the carrying capacity of distribution networks from three dimensions:safety margin,regulation capability,and investment efficiency.In addition,the Zonotope flexibility envelope technique is introduced to aggregate and model massive EVs, accurately characterizing the charge-discharge regulation boundaries of EV clusters.The Conditional Value-at-Risk（CVaR）is incorporated into the objective function to quantify extreme risks such as voltage violations,achieving a balance between economy and robustness.Through the coordinated scheduling of PV-ESS systems and EV clusters,the peak shaving and valley filling capabilities of energy storage and the flexible load regulation of EVs are utilized to significantly improve the safe and economic operation performance of distribution networks while ensuring charging demands.
• Select
  Research on HSSA-Optimized Multimodal Short-Term Load Forecasting Based on Electricity Spot Market Modeling

  WANG Yijun, WAN Shangjin, YOU Xiangyu, HOU Yuxuan

  2026, 46(3): 62-71. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0062-11

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Against the backdrop of the electricity spot market, the short-term load forecasting accuracy is limited owing to load variations affected by electricity prices, user behaviors and other factors. To address this issue, this paper proposes a multi-modal short-term load forecasting model optimized by a Hybrid Sparrow Search Algorithm (HSSA). A dual-branch architecture consisting of the Multi-Scale Temporal Convolutional Network (MS-TCN) and Quantum Phase Encoding-Transformer (QPE-Transformer) is constructed to hierarchically capture the short-term fluctuations and long-term dependencies of loads. A dynamic attention stacking module is designed to strengthen feature fusion during electricity price-sensitive periods and at load peak-valley fluctuations. Furthermore, the Hybrid Sparrow Search Algorithm is adopted to conduct global optimization of model hyperparameters. Relying on three mechanisms including hierarchical extraction of multi-modal features, real-time calibration of dynamic weights and intelligent optimization of hyperparameters, the model establishes a technical framework of “feature decoupling-dynamic fusion-global optimization”. Experimental results demonstrate that the proposed model can effectively cope with complex fluctuation scenarios in the electricity spot market and achieve a significant improvement in forecasting accuracy, which provides reliable technical support for real-time power grid dispatching and market supply-demand balance analysis.
• Select
  Robust Control of LCL Grid-connected Inverter Based on LMI Under Weak Grid Conditions

  MA Yukai, JIN Guobin, YANG Mingcheng, XIE Fei, LI Shuang, ZHOU Hailong

  2026, 46(3): 72-78. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0072-07

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Aiming at the problem that the stability of LCL grid-connected inverter system decreases due to impedance change and harmonic interference in weak grid, a state space model of three-phase LCL grid-connected inverter is established firstly, and several Quasi-Proportional Resonances (QPR) are introduced to ensure the non-static error output. A Linear Parameter-Varying (LPV) model of the system has been obtained. Then on the basis of LPV model, Linear Matrix Inequalities (LMI) was used to design state feedback controller. On this basis, voltage gain of Point of Common Coupling (PCC) was introduced, the harmonic voltage disturbance ability of the controller is further improved. The simulation results show that the designed robust controller still meets the requirements of the system for tracking effect and harmonic suppression under the condition of impedance changes and external interference.
• Select
  Operation Control and Voltage Support Strategies for Virtual Synchronous Generators under Grid Faults

  LIU Hongpeng, WANG Zhixing, GAO Hao, XU Ziqi, ZHANG Xueguang

  2026, 46(3): 79-90. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0079-12

  Abstract ( ) Download PDF ( )   Knowledge map   Save

   To address the challenges faced by Virtual Synchronous Generators (VSG), such as power-angle instability, fault current impact, and voltage recovery requirements during grid voltage dips, this paper proposes a Low Voltage Ride Through (LVRT) control strategy. The core of this strategy lies in: jointly regulating the active and reactive power reference values of the VSG during faults by dynamically adjusting the active power reference value to effectively suppress the virtual rotor angular velocity and maintain the power-angle stability of the VSG; while rapidly adjusting the reactive power reference value based on the depth of the grid voltage dip to provide active voltage support and assist in fault voltage recovery. To further suppress the peak fault current, a virtual impedance component is introduced in the strategy. This control strategy ensures that the VSG can simultaneously meet the grid connection standards for stable operation, reactive power support, and current safety during faults and the transient process after fault clearance. Finally, simulation experiments validate the effectiveness and robustness of the proposed control strategy, demonstrating that it significantly enhances the fault ride-through capability and system support performance of the VSG under grid fault conditions.
• Select
  Accurate Analytical Solution and Parametric Study on Local Buckling of Main and Diagonal Angle Stel Components in Transmission Towers

  SONG Dongrui, TANG Xiaocheng, SUN Zhiwei, HAN Dong, GUAN Xiaozhuo, LI Huashun

  2026, 46(3): 91-97. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0091-07

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Angle steel serves as a core load-bearing component of the main/diagonal members in transmission towers, and its local buckling behavior of angle steel directly affects the bearing capacity and safety of the tower structure, which is key issue in power engineering design. Based on the two-dimensional improved Fourier series method, this study conducts analytical solution research on the local buckling problem of angle steel under practical engineering scenarios. In the solving procedure, after performing Stokes' transformation on the selected trial function, the local buckling problem of angle steel can be converted into solving a set of linear algebraic equations, which significantly reduces the difficulty of theoretical solution during the engineering design phase and provides an efficient analysis tool for on-site engineers. Through numerical example comparisons, the critical load and buckling mode obtained in this study match well with existing analytical results and numerical simulation results, demonstrating the reliability of the proposed method in practical engineering applications. Based on the obtained analytical solutions, a quantitative analysis is further carried out on the commonly used angle steel section parameters in engineering, clarifying the quantitative influence laws of key geometric parameters such as aspect ratio and width-thickness ratio on the local buckling resistance of angle steel. This research provides theoretical support for the section selection, dimension optimization, and buckling-resistant design of angle steel used in the main and diagonal members of transmission towers.
• Select
  Research on Harmonic Characteristics and Suppression Strategies of High-Penetration Wind Power and Electrified Railway Load Grid-Connected System

  ZHANG Huaixun, LI Zhuoxia, ZHONG Cheng, ZHOU Ming, CHEN Zewen, LIU Guangyao, LIN Zeke

  2026, 46(3): 98-103. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0098-07

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  In the context of the transition to new energy and the electrification of transportation,the large-scale integration of wind power and electrified railways into the power grid has become a development trend.However,the combination of the characteristics of the two can easily lead to electrical energy quality issues such as harmonic distortion and negative sequence imbalance.This paper studies the impact of the joint connection of wind power and electrified railways on the electrical energy quality of the power grid.Using the harmonic state space method,a simulation model for the wind farm,the public power grid,and the electrified railway is established.The electrical energy quality indicators at the common connection point under different wind power penetration rates,load rates,and load imbalance degrees are analyzed. A harmonic and negative sequence collaborative control strategy is proposed. The research shows that when both the wind power penetration rate and the load imbalance degree act together,the voltage harmonic distortion rate is compounded and amplified. The greater the wind power penetration rate, the slightly higher the negative sequence voltage imbalance degree and the voltage harmonic distortion rate curve,indicating that the integration of wind power will exacerbate voltage imbalance.At the same time,the imbalance of the load aggravates the damage to the electrical energy quality caused by the integration of wind power.Through simulation and comparison of the electrical energy quality indicators under different operating conditions,the effectiveness of the proposed harmonic and negative sequence collaborative control strategy is verified.
• Select
  Short-time Trapping Efficiency Analysis of a Lift Trap with a Curved Baseplate in a Circuit Breaker with a Closed Resistance

  LIU Shili, HUANG Xianzhuang, LIU Qingda, ZHANG Fei

  2026, 46(3): 104-110. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0104-07

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Circuit breakers with closing resistors are prone to shedding particles in parallel with their breakers during operation.Particles are driven by electric field forces to initiate partial discharges that threaten reliable circuit breaker operation.Unlike the long-term stable operation of a GIL,the frequent switching of the AC filter in a converter station leads to intermittent,and short-term operation of the circuit breaker closing resistor,for which a curved base plate lift-up particulate trap structure is proposed.The results show that the curved-bottom lifting trap can change the direction of movement of particles and prolong the escape time,It can effectively suppress escape behavior and provide better capture efficiency at small time scales.
• Select
  Economic Optimization Scheduling of Active Distribution Network Time-Sharing with Electric Heating Load

  Fan Yuqing, Zhang Liwei

  2026, 46(3): 111-120. https://doi.org/10.19718/j.issn.1005-2992.2026-03-0111-10

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  To solve the problem of large load peak-valley difference in active distribution network (ADN) with high penetration rate of new energy,a time-sharing economic optimal dispatch method of ADN with electric heating load is proposed.Firstly,based on the first-order equivalent thermal parameter model and the state queue model,the formula for calculating the aggregate power of the electric heating load group is given,and the adjusted power of the electric heating load group is calculated;Secondly,considering wind power,photovoltaic,micro gas turbine,energy storage system and electric heating load as schedulable resources,an ADN time-sharing economic optimal scheduling model with the minimum comprehensive scheduling cost as the optimization objective is constructed;Then,the model is transformed into a second-order cone programming model by convex optimization technology,and the converted model is solved by using the YALMIP toolbox to call the CPLEX solver with better computational performance;Finally,the improved IEEE33-bus distribution network system verifies that the model can effectively improve the new energy absorption capacity of the distribution system,reduce the peak-valley load difference,and improve the economy of the distribution network operation.