最新刊期
- 摘要：Temperature-rise prediction of the main conductive system of high-voltage DC disconnectors is of great significance for equipment thermal design, hotspot identification, and condition assessment. In this paper, a graded thermal network modeling method oriented toward complex thermal behavior identification is proposed. Based on FEM electro-thermal coupling results, the regional thermal behavior complexity is identified in terms of local heat-source concentration, temperature-rise gradient enhancement, and heat-flow path deflection, thereby enabling graded construction of the thermal network model for the main conductive system. Taking a 408 kV disconnector as the study object, finite-element electro-thermal coupling analysis is carried out to extract the characteristics of current density, temperature field, and heat-flow field. Regional complexity indices are established, and explicit contact thermal resistance, equivalent-parameter thermal networks, and lumped-parameter thermal networks are adopted for graded modeling of different regions. Two current-carrying conditions, namely 5 000A and 8 000A, are selected for comparison, and an 8 000 A DC temperature-rise test is conducted. The results show that the contact–finger interface is the key region for temperature-rise prediction. The graded thermal network is consistent with FEM results in terms of temperature-rise trend and hotspot location. For 15 experimental measurement points, the average relative error is approximately 2.3%, and the maximum relative error is approximately 3.1%, indicating high prediction accuracy. The proposed method can improve the local temperature-rise characterization capability of key contact regions while controlling the scale of the thermal network, thereby providing a reference for temperature-rise prediction and thermal design optimization of high-voltage DC disconnectors.关键词：High-voltage DC disconnector;temperature-rise prediction;complex thermal behavior;graded thermal network;FEM0|0|0更新时间：2026-09-07
-
Research on multi-objective control strategy of power conversion system under asymmetric voltage sag
摘要：When asymmetrical voltage dips occur in the power grid, it will adversely affect the safe grid integration of energy storage. A multi-objective control strategy for power conversion system (PCS) under asymmetrical voltage dips is proposed. The strategy takes supporting the grid-connected point voltage to a safe range as the primary objective, and designs two sets of voltage reference values, which can be selected according to the magnitude of the actual negative sequence voltage. Subsequently, the optimal voltage support method is determined by the Lagrange multiplier method, and the constraints of PCS output current and active power fluctuation are considered to ensure the safe operation of the grid-connected system during faults. When supporting the voltage at the grid-connected point to a safe range, the voltage unbalance is further suppressed according to the PCS residual capacity margin to improve the voltage quality. Finally, the effectiveness of the proposed strategy is verified by MATLAB/Simulink. When the voltage transient degree is small, it can not only ensure the voltage support effect, but also better limit the active power fluctuation; when the voltage transient degree is large, it can give the grid-connected system more abundant non-off-grid operation time.关键词：power conversion system;Grid-connected energy storage;asymmetric voltage sag;Voltage Support;active power fluctuation suppression1|0|0更新时间：2026-09-02 - 摘要：With the increasing voltage level of vacuum circuit breakers, the dynamic contact load on the cam of the spring operating mechanism increases, raising the risk of contact fatigue failure. Conventional cam design usually treats contact strength as a post-checking index and lacks profile optimization driven by dynamic contact stress feedback. To address this issue, a closing profile optimization method based on co-simulation and Hertz contact theory is proposed. NURBS control points are used as design variables, the maximum dynamic contact stress is minimized, and contact strength criteria and mechanism motion constraints are imposed. The optimization model is solved by sequential quadratic programming within an Adams-MATLAB co-simulation framework. A 252 kV vacuum circuit breaker is used for verification. The results show that the maximum contact stress decreases from 1287.8 MPa to 1019.6 MPa, and the maximum normal contact force decreases from 87.1 kN to 72.8 kN, with reductions of 20.8% and 16.4%, respectively. The maximum operating pressure angle and follower angular acceleration are also reduced, indicating improved cam transmission characteristics. After 10 000 mechanical endurance cycles, no pitting, peeling, or abnormal wear is observed on the cam working surface. The proposed method reduces the dynamic contact stress while ensuring closing mechanical characteristics, providing a reference for reliability optimization of key transmission components in high-voltage vacuum circuit breakers.关键词：vacuum circuit breaker;cam;Co-simulation;Hertz Contact Theory;SQP4|4|0更新时间：2026-08-27
- 摘要：Potential transformers are frequently exposed to harsh operating conditions. Under various overstresses, local inter-turn short circuit faults may occur in the winding insulation, which affects the metering performance and operation safety of the equipment. However, the existing research on the leakage flux distortion and local overheating caused by the local inter-turn short circuit fault of the potential transformer is relatively insufficient, and it is difficult to reveal the influence of the short circuit position and the number of turns on the distribution of the magnetic-thermal field. In addition, the existing fault identification features mostly focus on single point amplitude or local state, which is difficult to describe the spatial difference of fault response, resulting in limited identification ability of different fault states. In order to solve the above problems, this paper proposes a method for identifying the inter-turn short circuit fault of potential transformer driven by magnetic-thermal axial distribution characteristics: Based on COMSOL finite element software, the field-circuit coupling and magnetic-thermal coupling simulation models are constructed, and the axial distribution sequences of leakage flux and temperature are extracted to construct the magnetic-thermal axial distribution morphological characteristics used to characterize the fault response. Furthermore, the Bayesian optimized extreme gradient boosting ( XGBoost ) model is used to classify and verify the single physical quantity feature and the magnetic-thermal fusion feature to evaluate the effectiveness of the feature system. The simulation results show that the short circuit position mainly affects the axial distribution area of the magnetic-thermal anomaly response, and the number of short circuit turns mainly affects the magnetic flux leakage amplitude and the temperature rise of the hot spot. The model identification effect is the best under the magnetic-thermal fusion feature input, and the accuracy rate is 96.15 %.关键词：potential transformer;Inter-Turn Short Circuit Fault;Magnetic-Thermal Coupling;Axial Distribution Characteristics;fault identification7|4|0更新时间：2026-08-25
- 摘要：The double pulse test is one of the important characterization tests for integrated gate-commutated thyristor (IGCT) and is commonly used to analyze and evaluate the device's dynamic performance. However, the stray inductance of the loop in double pulse test generates an induced voltage when the IGCT is turned off, which together with the bus voltage results in an overvoltage on IGCT. Excessive stray inductance can lead to errors in the dynamic characteristic measurement, or even damage the device.To address this issue, this paper proposes a novel approach to reduce the loop stray inductance and optimize the IGCT double pulse test platform using particle swarm optimization (PSO) algorithm. First, the three-dimensional models of components in the double pulse test loop are developed, including the IGCT fixture, the diode fixture and the clamp capacitor. Their spatial configuration is parameterized for representation. The optimization range for spatial configuration parameters is defined, considering the constraints imposed by connection sequence and insulation spacing. Then, the PSO algorithm is employed to find the optimal spatial configuration of platform with the minimum value of stray inductance. Finally, the circuit simulation verifies that the proposed optimization approach can effectively suppress the overvoltage amplitude when the IGCT is turned off. The physical double pulse test platform is constructed based on the optimized spatial configuration, and the experiments are conducted. Comparing the experimental results and simulation results, it can be observed that the discrepancy between the two remains consistently within a small range, further demonstrating the feasibility of the proposed approach.关键词：IGCT;Double pulse test platform;stray inductance;Spatial configuration;PSO algorithm11|3|0更新时间：2026-08-13
- 摘要：To address the insufficient prediction accuracy of traditional deterministic simulation methods for Very Fast Transient Overvoltage (VFTO), which stems from ignoring the stochasticity of disconnector breakdown voltage, this paper proposes a high-precision probabilistic simulation method driven by measured data. Based on the measured distribution characteristics of breakdown voltage in a 1100 kV gas-insulated switchgear (GIS) disconnector, a dynamic time-slice division and segmented statistical modeling approach is developed to construct a breakdown criterion of "distribution interval bounds + random disturbance." This criterion is embedded into ATP-EMTP to enable efficient simulation of stochastic reignition processes. Experimental results demonstrate that, compared to traditional linear models (Linear Model I and Linear Model II), the proposed method reduces the residual sum of squares (RSS) for VFTO peak prediction from 0.0255 and 0.0517 to 0.0049, and decreases the root mean square error (RMSE) from 0.053 and 0.075 to 0.023. Under full phase angles, the maximum VFTO peak error is significantly reduced from 4.409% and 4.773% to 0.73%, achieving an 86.7% improvement in consistency with measured data. By quantifying the enhancement in simulation accuracy through stochastic modeling, this study reveals the over-conservative design flaws of traditional deterministic models caused by neglecting dispersion, providing a high-confidence simulation tool for insulation optimization in ultra-high voltage equipment.关键词：Very Fast Transient Overvoltage (VFTO);Ultra-High Voltage Gas-Insulated Switchgear (GIS);disconnector;Stochastic Breakdown Voltage;Probabilistic Simulation2|9|0更新时间：2026-08-13
- 摘要：Insulation defects in gas-insulated switchgear (GIS) are the primary cause of insulation performance deterioration and safety accidents. The conventional ultrasonic testing method is limited by the response frequency band of the piezoelectric sensors and the resonance feature of the GIS metal shell. Broadband partial discharge ultrasound signals cannot be obtained accurately by this method, affecting the effectiveness of insulation defect detection. investigate. In this study, the acoustic emission characteristics of typical insulation defects in GIS with broadband acoustic sensors, including metal spike defect, free particle defect, floating potential defect and air-bubble defect inside epoxy, were investigated. The results show that acoustic signals generated by metal tip defect and free particle defect are cluster-shaped, with a longer duration and a lower peak frequency, while that of floating potential defect is pulse-shaped, with a short duration and a higher peak frequency. Besides, it is noted that the peak frequency and amplitude of insulation defect sound signals are strongly correlated with defect size. The equivalent time equivalent frequency method can distinguish the four types of defect sound signals effectively. The research results have a guiding role in the detection and diagnosis of partial discharge acoustic signals of typical defects in GIS.关键词：GIS;metal spike;free metal particle;floating potential;air-bubble inside epoxy;acoustic emission;optical microphone21|18|0更新时间：2026-07-21
- 摘要：To enhance the detection efficiency and positioning accuracy of high-voltage cable outer sheath grounding fault, and address the limitations of traditional positioning methods that rely on power outage operation, positioning delay and high implementation complexity, a live positioning method based on sheath circulating current characteristics is proposed. Taking single-ended protective grounding and cross-connected grounding as examples, the formation mechanism of sheath current is analyzed theoretically, and the electromagnetic transient model of three-phase single-core cable is built and validated based on PSCAD / EMTDC. By simulating the dynamic process of the outer sheath grounding fault, the distribution characteristics of the sheath current varying with the fault location are analyzed. A polynomial fitting method is employed to establish the sheath current-fault location characteristic equation, which, combined with current inversion at the direct grounding end, enables real-time online fault location. Engineering case studies have verified the feasibility of the proposed method. For single-end grounding systems, fault location can be achieved through single-point current measurement, while cross-connected systems require synchronous current monitoring at both ends to identify fault sections and determine precise locations, without power outage and low operation complexity. The proposed method provides an effective technical reference for on-line detection and rapid location of high-voltage cable outer sheath grounding fault, and has important engineering application value.关键词：high-voltage cable;sheath circulation;grounding fault;characteristic equation;live positioning15|9|0更新时间：2026-07-17
- 摘要：To investigate the correlation between gas generation and capacitive screen micro-discharge in oil-immersed upright current transformers, failure analysis was performed on two out of three 220kV current transformers (same model) that experienced bellows expansion tank cover failures in a regional power grid. The faulty units underwent insulation resistance measurement, dielectric loss tangent (tanδ) and capacitance testing, and dissolved gas analysis (DGA). Results revealed that hydrogen (H2), total hydrocarbon content, and ethylene (C2H4) concentrations in the insulating oil exceeded warning thresholds, accompanied by elevated dielectric loss values. Post-disassembly inspections identified significant wrinkling in the insulation paper surrounding the primary winding and across capacitive screen layers, indicating critical manufacturing defects. Using COMSOL Multiphysics 6.3, finite element models were developed to simulate the primary winding’s external insulation structure, oil-paper interfaces, and wrinkled air gaps. Partial discharge processes under these defect conditions were analyzed, confirming that insulation irregularities and capacitive screen deformation induced localized electric field concentrations. Combined with experimental and dissection findings, the root cause was attributed to progressive micro-discharges at wrinkled regions, exacerbated by degraded oil insulation properties, ultimately leading to gas accumulation and structural failure.关键词：Current Transformer (CT);dissection;Dissolved Gas Analysis (DGA);finite element;Partial Discharge (PD)24|41|0更新时间：2026-07-06
- 摘要：Under the monitoring framework for power system state perception, in response to the problem that the multi-modal and wide-band characteristics of the electric field signals in the power system are prone to cause spectral aliasing, which affects the accuracy of subsequent equipment state identification and risk assessment, a quantum-enhanced multi-modal electric field monitoring method is proposed. This method collects current signals based on the diamond NV color center fiber quantum sensor, and adopts a meta-learning enhanced microwave frequency modulation strategy. Through multi-condition prior learning and rapid adaptation mechanisms, it dynamically optimizes the modulation frequency offset and modulation frequency to ensure that the quantum sensing always operates at the optimal sensitivity point. Further, the corrected current sequence is input into the bidirectional long short-term memory network, and the forward and backward temporal feature extraction capabilities are utilized to construct dynamic assessment models for normal operation, alert, emergency, and fault states. Tests show that this modulation method increases the current sensing sensitivity from 86.9 A/GHz to 1146.2 A/GHz, an increase of approximately 13.2 times. Under a 0.5 high proportion noise interference, the model's recognition results for each state type are completely consistent with the preset state, and the perception difference in the full current range is lower than 0.1, verifying the robustness and accuracy of the proposed method in high-noise environments.关键词：power system;State perception;Dynamic evaluation;Meta learning enhancement;Quantum sensing monitoring;NV color center current transformer;Bidirectional Long Short Term Memory Network26|2|0更新时间：2026-05-25
- 摘要：The circuit breaker of the converter station filter group has the characteristics of frequent operation，harsh operating environment，and high failure rate，and its reliability urgently needs to be improved. This article proposes a reliability centered maintenance strategy for filter group circuit breakers. Firstly，the fault characteristics of filter group circuit breakers are analyzed to obtain equipment maintenance project requirements mainly focused on arc extinguishing chamber defects; Secondly，establish an optimization model for the operation and maintenance strategy of the converter station filter group circuit breaker with the lowest equipment operation risk as the optimization objective，and calculate the benefits after adding an arc extinguishing chamber maintenance project; Finally，based on the collected on-site operation data of a specific type of circuit breaker，it is proposed that conducting an internal inspection of the opening and closing of this type of circuit breaker after 205 opening and closing operations has the best efficiency. Based on the research results of this article，a reliability centered optimization method for the maintenance strategy of filter group circuit breakers has been summarized，which has certain reference significance for improving the maintenance strategy of filter group circuit breakers in converter stations in the future.关键词：filter group circuit breaker;maintenance strategy;reliability;optimized model;reliability centered maintenance26|55|0更新时间：2026-05-19
- 摘要：As the core components of circuit breaker, electrical contacts need to have excellent electrical conductivity, mechanical properties and anti-melting properties. CuCr alloys are widely used because of high strength and high conductivity, but there are defects such as brittleness enhancement at high temperatures. By constructing a model of CuCrX (X=C, Si) replacement solid solution with different Cr concentrations, the lattice parameters, thermodynamic stability, elastic constants and electronic structure of the alloys were analyzed using CASTEP software. The results show that C doping significantly affects the lattice constant, the lattice distortion increases with the increase of Cr concentration, and significantly improves the thermodynamic stability, but reduces the modulus of elasticity, enhances the brittleness, and its elastic change is favorable to improve the resistance to fusion welding. Si doping has a weak effect on the lattice and enhances the elastic modulus, CuCr₀.₁₆Si₀.₀₃ has the optimal integrated elastic strength, while enhancing brittleness to improve fusion welding resistance. The electronic structure shows that both C and Si doping enhance the electrical conductivity, and the theoretical electrical conductivity of CuCrC is stronger. In terms of bonding characteristics, CuCrC hybridization plays a significant role, and Si-Cr hybridization in CuCrSi enhances the bonding energy and deformation resistance. This study provides a theoretical basis for the optimization of the properties of CuCr-based contact alloys.关键词：First-principles;C/Si single doping;Thermodynamic stability;Elastic properties;Electronic structure49|48|0更新时间：2026-04-24
- 摘要：Ultrasonic signal measurement is one of the effective on-line detection methods of partial discharge in gas insulated equipment. The optical fiber interferometer sensing system has the advantages of high sensitivity and anti-electromagnetic interference, so it is suitable for partial discharge ultrasonic detection, but it is easily disturbed by environmental noise and affects the signal measurement. In this paper, firstly, the sensing principle of ultrasonic signal of optical fiber interferometer is defined, the phase change of sensing light under the action of ultrasonic signal is quantified, and the cause of phase fading is analyzed, secondly, the methods of high frequency carrier modulation and downconversion demodulation are introduced into the Michelson interferometer topology, which overcomes the influence of environmental noise on the sensing system and realizes the real-time demodulation of ultrasonic signal. Then, an ultrasonic measurement platform is built in the laboratory to verify the accuracy of the down conversion demodulation method. Finally, based on the research system, field testing experiments are carried out on the GIS equipment of a 1000 kV substation, the partial discharge ultrasonic waveforms were successfully captured. The ultrasonic sensing system proposed in this paper offers a sensitive and reliable approach for detecting partial discharge in gas-insulated equipment.关键词：GIS;partial discharge;ultrasonic sensing;optical fiber interferometer;frequency down conversion32|46|0更新时间：2026-04-09
- 摘要：Composite insulators are widely used in power grids due to their excellent insulation performance and mechanical strength. However, the hidden defects such as debonding and delamination caused by the sheath-core interface are the main causes of malignant faults such as broken strings and breakdown. The existing detection methods are difficult to meet the high-sensitivity detection requirements of such defects. In this paper, a diagnosis method of interface defects of composite insulators based on nonlinear ultrasound is proposed. The composite insulator samples without defects, debonding and delamination were taken as the research objects. The nonlinear ultrasonic test platform was built and the ultrasonic echo signals were collected. The discrete Fourier transform-least squares ( DFT-LS ) method was introduced to extract the nonlinear characteristics in frequency domain, and the interface state was characterized by combining the statistical characteristics in time domain. A defect diagnosis model based on Improved Grey Wolf Optimizer-Support Vector Machine ( IGWO-SVM ) is established. The results show that the DFT-LS algorithm significantly reduces the calculation error of relative nonlinear coefficients in high noise environment and effectively improves the robustness of feature extraction. The IGWO-SVM model is superior to GWO and other similar optimization algorithms in terms of convergence speed and recognition accuracy. The model recognition accuracy reaches 96.7 %, which can realize effective detection and intelligent diagnosis of interface defects of composite insulators.关键词：composite insulator;Interface defect;Nonlinear ultrasonics;Improved Grey Wolf Optimizer;support vector machine41|5|0更新时间：2026-04-08
- 摘要：GIS has the advantages of small space and high insulation performance, and is widely used in power grid. However, the movement of free metal particles in its interior will reduce its insulation stability, leaving hidden dangers for the safe and reliable operation of the equipment. Firstly, the motion behavior of free metal particles and the change of take-off voltage under different working conditions are studied. The results show that vibration excitation can broaden the motion range of metal particles and effectively reduce their take-off voltage by 29.8%, and then three typical motion modes of the particles are established. Secondly, pulse current method was used to measure and analyze the discharge characteristics of metal particles under vibration excitation. The results showed that the application of vibration increased the dispersion of PRPD spectra of metal particles to a certain extent, and the maximum discharge of small size and low density metal particles significantly increased, and the maximum discharge of 0.5mm aluminum particles increased by 146%. And after reaching a certain voltage threshold rapidly increase, easy to cause breakdown; The mechanical vibration excitation has little effect on the partial discharge characteristics of heavy particles such as 1mm copper particles. The research results can provide theoretical support for the detection and analysis of free metal particles in GIS active circuit breakers in practical engineering.关键词：GIS;metal particles;mechanical vibration;partial discharge;Motion law;Take-off voltage;pulse current method;discharge characteristic49|38|0更新时间：2026-04-03
- 摘要：On-site testing of gas-insulated switchgear (GIS) is the final safeguard to ensure safe operation after commissioning. At present, AC withstand voltage tests are widely used in engineering practice; however, significant discrepancies exist between the test conditions and the actual “high-voltage–high-current” combined stresses experienced during GIS operation. As a result, multiple cases have been reported in which severe insulation breakdown occurred shortly after the equipment had successfully passed the AC withstand test. To address the limitations of existing test methods, this paper proposes a combined voltage–current on-site test for GIS, aiming to enhance the detection of insulation defects before commissioning and improve the operational reliability of GIS. A technical approach for applying high voltage and large current simultaneously under field conditions is analyzed, and a novel on-site test method based on resonant voltage and induced current combined excitation is proposed, together with a complete testing procedure. Furthermore, an induction-type high-current test device suitable for field application is developed, and the combined voltage–current test is conducted for the first time on an actual GIS during its handover testing. The results indicate that when applying a large induced current to the GIS internal conductor, conductive loops involving the GIS enclosure, enclosure grounding, and substation grounding grid must be avoided. By coordinating two low-power excitation sources, the proposed method can safely and effectively realize combined voltage–current on-site testing, offering a practical solution to improve the authenticity and effectiveness of GIS handover tests.关键词：GIS;On-site testing;Combined voltage-current test;Handover test;partial discharge;Effectiveness;Induction-type current booster50|55|0更新时间：2026-03-24
- 摘要：At present, with the vigorous development of power Internet of Things, wireless sensor networks composed of various sensor terminal devices and small wireless base stations have become an important part of the perception layer of power transmission Internet of things. However, the continuous and reliable power supply of distributed sensor nodes in outdoor transmission lines is a key problem that needs to be solved at present, and the environmental micro-energy acquisition technology provides an effective solution for the self-driven power supply of sensor nodes in transmission lines. To address the continuous power supply of distributed monitoring nodes for transmission lines in outdoor environments, this paper designs an electromagnetic-piezoelectric hybrid energy-harvesting system based on a suspended antivibration structure, targeting both the ambient wind energy around transmission lines and the conductor’s micro-aeolian vibration energy. By integrating the suspension structure of the wind cup with the anti-vibration structure of the wire, a wind-driven electromagnetic power generation unit and a vibration-driven vibration damping piezoelectric power generation unit are designed. Furthermore, a power management strategy with undervoltage locking function is proposed to provide standardized DC output for wireless sensor nodes. Compared with the traditional rectifier charging power supply mode, the power management circuit reduces the power consumption by 94.7%, which can realize efficient and continuous power supply for sensor nodes. The electromagnetic-piezoelectric composite energy acquisition system designed in this paper provides a new solution to solve the problem of green, continuous and reliable power supply of online monitoring sensor nodes of transmission lines.关键词：transmission lines;Self-powered;Suspension structure;Antivibration structure;Electromagnetic power generation;Piezoelectric power generation;Power management23|52|0更新时间：2026-03-23
- 摘要：The ash particles produced by the burning of vegetation after a wildfire can adhere to the surface of insulators. As conductive contaminants, these ash particles can affect the insulation performance of insulators. In this paper, ash particles from the burning of vegetation were obtained through a burn loss test, and flashover tests were conducted on three-string XWP2-160 porcelain insulators to study the influence of ash particle concentration (TCD), initial salt density (ISD), and ash attachment area (ST) on the flashover voltage and leakage current of insulators. The research shows that as TCD increases, the flashover voltage of contaminated insulators decreases approximately linearly, and the peak leakage current increases as a power function with a positive exponent. Compared with the situation without ash attachment, the flashover voltage decreases by 19.8% to 31.5%, and the peak leakage current increases by 16.6% to 44.4%. Under the condition of ash attachment, as ISD increases, the flashover voltage of insulators decreases as a power function with a negative exponent, and the influence of TCD changes on the flashover voltage and peak leakage current of contaminated insulators gradually decreases. As ST increases, the flashover voltage gradually decreases, and the peak leakage current gradually increases. Both show a trend of "initially gentle - significantly in the middle - gradually slowing down in the later stage". The research results can provide a reference for the construction of risk assessment models and operation and maintenance strategies for insulators on transmission lines after wildfire.关键词：porcelain insulator;TCD;flashover voltage;Leakage current peak value;ISD;ST41|28|0更新时间：2026-03-18
- 摘要：In high-voltage vacuum circuit breakers, the design of contact structure plays a key role in the magnetic field distribution characteristics as well as the suppression of dynamic instability triggered by electrodynamic repulsive force on the operation of circuit breakers. To this end, a 3D finite element model of the cup-shaped longitudinal magnetic contact is established in this study, and by fusing support vector regression (SVR) and sparrow search algorithm (SSA), six key structural parameters, namely, contact piece slotted length L1, width L2, radial rotation angle θ1, cup seat thickness W, slant slot up and down rotation angle θ2, and slant slot height H, are used as inputs, and the electrodynamic repulsive force F, the center plane at the peak of the current The SSA-SVR high-precision prediction model is constructed by taking the electric repulsive force F, the effective magnetic field occupation ratio S at the peak current and the hysteresis time t0 at the center point when crossing zero as outputs, respectively; finally, the Pareto front is generated by NSGA-Ⅲ, and the parameter optimization is realized by combining with the TOPSIS method. The results show that: compared with the initial data, the parameter optimization reduces F from 169.91N to 105.29N,S increases from 53.31% to 63.07%,and t0 reduces from 1.0233ms to 0.7899ms,which significantly improves the opening performance and operation stability of the contact.关键词：vacuum interrupter contacts;electrodynamic repulsion;magnetic field characteristics;SSA-SVR;NSGA-III;TOPSIS63|80|0更新时间：2026-02-11
- 摘要：To address the problem of low mechanical fault diagnosis rate of electromagnetic repulsive actuator mechanism, the paper proposes a fault diagnosis method based on the Improved Dung Beetle Optimization (IDBO)algorithm to optimize the Gated Recurrent Unit (GRU)neural network ,which firstly adopts the Energy Entropy of Maximum Singular Value (EEMSE) to construct the feature vector matrix for feature extraction of the slamming vibration signals of the mechanism, and then utilizes the Lasso regression coefficient and the Pearson correlation coefficient to reduce the dimension of the feature matrix.Then, Chebyshev chaotic mapping, golden sine strategy, and dynamic weight coefficients of position update are introduced to improve the traditional Dung Beetle Optimization (DBO)algorithm, and the hyperparameter optimization of GRU using IDBO is used to establish the fault diagnosis model of electromagnetic repulsive actuator mechanism based on IDBO-GRU.Finally, the experimental platform is built to verify the model, and the results show that the diagnostic accuracy of the established fault diagnosis model reaches 96.88%, which is higher in accuracy and better in stability than other diagnostic models, and provides a new diagnostic model for the fault diagnosis of the electromagnetic repulsive actuator.关键词：Electromagnetic repulsive actuatorsr;Improving the Dung Beetle Optimization Algorithm;Neural network of gated recurrent units;Feature extraction and screening;fault diagnosis47|83|0更新时间：2026-02-11
- Technical support is provided by Beijing Founder electronics co., LTD 陕ICP备05005324号-4
陕公网安备 61010402000197 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0