This article presents an ultrasonic intelligent diagnostic method for internal defects in ultrahigh voltage basin insulators using an improved YOLOv8 model. This approach addresses the limitations of manual interpretation in traditional ultrasonic inspections by enabling automatic analysis and intelligent defect identification. The proposed method incorporates a deeper network architecture with optimized loss functions and adaptive feature extraction mechanisms, significantly enhancing adaptability in complex scenarios. Compared to YOLOv5 and YOLOv8n, the improved model achieves a 7% and 3% increase in ${F}1$ score, respectively, along with a 13.6% and 10.3% improvement in mAP@0.5. Experimental validation on standard defect samples and a three-phase common box basin insulator demonstrates the model's capability to detect 1 mm-scale bubbles and cracks with less than 2% localization error, over 80% diagnostic confidence, and 93.5% consistency with industrial Digital Radiography (DR) inspections, providing a robust and intelligent solution for defect detection in critical power system components.

Aiming at the 500 kV oil-paper insulated submarine cable terminal, an electric field simulation platform is established to verify the electric field distributions of structures with and without an oil gap under AC 740 kV, DC 1030 kV and lightning impulse 1550 kV. For the structure without oil gap, the radial electric field in the capacitance cone is uniform, with a maximum field strength of about 24.5 kV/mm under DC and about 15.6 kV/mm under AC; the maximum field strength on the conductor surface is about 54 kV/mm, and on the grading ring about 4.1 kV/mm. The oil gap is caused by the geometric discontinuity of the electrode edges and the winding process. Under AC, the field strength in the oil gap can reach 8.8 kV/mm, but the insulation margin is still satisfied. The influence of 90 °C hot state on the electric field is less than 10%. In the conclusion, preventive measures such as optimizing electrode edges and controlling winding tension are proposed.

To address the growing overheating risk of transformers—where excessive temperatures shorten insulation service life and trigger breakdowns, endangering the safe operation of power systems—this study proposes an online reconstruction technique for the transient temperature field of oil-immersed transformer windings, integrating Gappy Proper Orthogonal Decomposition (Gappy-POD) with discrete sensor data. First, high-fidelity numerical simulation data under multiple startup conditions are used to extract POD standard orthogonal bases that capture the dominant features of the temperature field. Second, a greedy algorithm is adopted to optimize sensor placement, enhancing the technique’s robustness against measurement errors. Finally, the full-domain temperature field is reconstructed by minimizing the error between measured and reconstructed values. Experimental validation with real sensor data shows that the reconstructed results are consistent with the measured values at verification points, with a per-step reconstruction time of approximately 0.015 s. This technique provides an effective solution for real-time transformer operation monitoring and early fault warning, while offering a reference for the digitalization of similar power equipment.

This paper investigates the distribution pattern of residual stress during the curing process of 550 kV tri-post insulators to address equipment failures caused by curing defects. Field experiments utilizing the Bragg grating method are conducted, incorporating coefficient calibration and data analysis for a systematic study. First, tensile and temperature tests are performed to calibrate the strain sensitivity coefficient and temperature sensitivity coefficient of the Bragg grating, establishing a foundation for accurate measurements. Subsequently, gratings and thermocouples are embedded at specified positions within the insulator mold to monitor temperature and strain changes throughout the pre-curing and post-curing stages, where the strain variations inside the insulator effectively reflect stress changes. The test results indicate that during the curing heating stage, the temperature distribution within the insulator is uniform. The epoxy composite material exhibits actual temperatures that exceed the set values due to curing exotherm, with the fastest heating and cooling rates observed in regions near the inserts. During the curing process, tensile strain and stress are recorded at measurement points close to the inserts, while contraction stress dominates in areas farther from the inserts. Demolding triggers abrupt changes in stress and strain, with the maximum contraction rate occurring near the central tube during the cooling stage. This study reveals the distribution characteristics of curing strain and the evolution of residual stress in 550 kV tri-post insulators, providing critical experimental data and technical references for optimizing the curing process and enhancing the mechanical performance and operational reliability of insulators.

Structural strength of transformer tank is a key factor to ensure the stable operation of transformer. To address the design verification needs of large transformer tanks, the stress-strain characteristics of transformer oil tank under positive pressure, negative pressure, oil pressure and temperature rise are simulated and analyzed according to the design verification requirements of large transformer tank. Combined with the actual operating conditions of transformer, the strain distribution of transformer oil tank under the comprehensive action of positive pressure, oil pressure and temperature rise are analyzed. The results show that there is significant local displacement in the central area of the side wall. To optimize the strain distribution, a quantitative analysis is performed on the influence of positional parameters of stiffeners reinforcements on structural strength. Four stiffen-er positions with substantial impact on deformation are selected as optimization variables. A surrogate model is constructed to optimize their positions, revealing that displacements of 23 mm, −45 mm, −70 mm, and − 120 mm for the four stiffeners reduce the maximum surface displacement by 6.1

When high-voltage DC equipment is subjected to polarity-reversing voltage, the insulation performance of its internal insulation material, epoxy resin-impregnated paper (RIP), is significantly affected by low-frequency polarization. To clarify the underlying mechanisms, this study analyzes the impact of the relationship between low-frequency dielectric relaxation time and polarity-reversing time on the dynamic changes of space charges at different temperatures based on the pulsed electro-acoustic (PEA) method. Furthermore, the influence of traps at the epoxy resin (EP)/RIP interface on interfacial charge accumulation is considered. By improving the existing bipolar carrier transport model, the dynamic characteristics of interfacial space charges in RIP under different polarity-reversing times are explored. The results show that at 313K, due to the low-frequency dielectric relaxation time constant exceeding 60s, the charge distribution is not significantly affected by the reversal time; At 333K and 353K, the reduction in dielectric relaxation time leads to a decrease in interfacial charge density and maximum field strength, with the peak field strength shifting towards the epoxy layer; At 373K, influenced by space charge dissipation and enhanced relaxation effect, the electrode surface charges and the maximum field strength in the epoxy layer initially decrease and then increase with increasing reversal time. The simulation results more explicitly characterize the interfacial polarization process between EP and RIP, and together with the experimental results, elucidate the mechanism of low-frequency polarization on the dynamic characteristics of space charges in RIP under polarity-reversing voltage. The research results may provide a theoretical basis and reference for the insulation design and optimization of high-voltage DC equipment.

The valve-side bushing of converter transformer is a critical component in both AC and DC power grids and plays an essential role in ensuring the safe and stable operation of power systems. Flashover at the gas–solid interface between the valve-side bushing core and SF₆ is closely related to surface charge accumulation. Understanding surface charge accumulation characteristics under DC voltage is therefore of great significance for the reliable operation of valve-side bushings. In this study, the isothermal surface potential decay method was employed to investigate the trap distribution of epoxy resin impregnated paper used in valve-side bushing cores at different temperatures. The trap distribution of epoxy resin impregnated paper samples coated with TiO2 was also measured. The DC creepage flashover performance of coated and uncoated samples in SF₆ was further examined. The experimental results indicate that, with increasing temperature, the shallow trap density increased while the deep trap density gradually decreased for both types of samples. Moreover, the TiO2 coating reduced the deep trap energy level of epoxy impregnated paper at temperatures above 70 °C. In cylindrical sample tests, the TiO2 coating altered the surface conductivity of epoxy-impregnated paper, thereby modifying the charge distribution at the gas–solid interface. In DC surface flashover tests conducted in SF6, the average flashover voltage of TiO2-coated samples was 66.47 kV, which was higher than that of epoxy resin impregnated paper sample’s 55.70 kV. This coating effectively enhanced the creepage flashover voltage and exhibited superior resistance to electrical tracking. These findings provide guidance for the design of surface coatings for valve-side bushing cores aimed at promoting charge dissipation and improving resistance to electrical tracking.

To address the low computational efficiency of multiphysics simulations for oil-immersed transformers, which limits real-time digital twin applications, a fast temperature field reconstruction method for transformer windings is proposed based on discrete temperature sensors and Gappy Proper Orthogonal Decomposition (Gappy POD). High-fidelity numerical simulations under multiple operating conditions are performed to construct a temperature snapshot set, from which dominant POD modes are extracted. Using a small number of sensor measurements, the global transient temperature field is reconstructed online. A S20-M-630 kVA distribution transformer is used as a case study. Results show that accurate temperature reconstruction can be achieved with only four sensors, with a global average relative error below 2% and a computation time of approximately 0.012 s, demonstrating the effectiveness of the proposed method for transformer digital twin applications.

ABSTRACT As a commonly used bushing current‐carrying structure in recent years, contactor‐type electrical connection parts operate in a complex environment of electric, thermal and chemical corrosion for extended periods. With the increase of ultra‐high voltage (UHV) transmission capacity, its operating environment faces more stringent challenges, and the stability of their operation is directly linked to the reliability of the UHV transformer equipment. This paper aims to investigate the mechanism behind contact failure by developing a contactor deterioration test platform in oil to study the deterioration and corrosion process of strap contactor in oil under 120°C environment. The study found that the sulphur element content on the exposed copper surface of the strap contactor was elevated from 0% to 2.05% in 28 days, and there was an obvious latency period for the sulphide corrosive reaction with the surface of the strap contactor. The electrical contact resistance of the strap contactor after the corrosion reaction was elevated from 110 to 251 μΩ, which became overheating failure and contact failure of electrical connection components of bushings.

With the rapid development of high-voltage DC (HVDC) power systems, accurate measurement of surface electrostatic potential on insulating components has become critical for electric field assessment and insulation reliability. This paper proposes an electrostatic potential sensor based on cantilever micro-vibration modulation, which employs piezoelectric actuators to drive high-frequency micro-vibration of cantilever-type shielding electrodes, converting the static electrostatic potential into an alternating induced charge signal. An electrostatic induction model is established to describe the sensing principle, and the influence of structural and operating parameters on sensitivity is analyzed. Multi-physics coupled simulations are conducted to optimize the cantilever geometry and modulation frequency, aiming to enhance modulation efficiency while maintaining a compact sensor structure. To validate the effectiveness of the proposed sensor, an electrostatic potential measurement platform for insulating components is constructed, obtaining response curves of the sensor at different potentials and establishing a compensation model for the working distance correction coefficient. The experimental results demonstrate that the sensor achieves a maximum measurement error of 0.92% and a linearity of 0.47% within the 1-10 kV range. Surface potential distribution measurements of a post insulator under DC voltage agreed well with simulation results, demonstrating the effectiveness and applicability of the proposed sensor for HVDC insulation monitoring.

With the continuous increase in power grid transmission capacity, the current load borne by the strap type electrical connection components within the valve-side bushings of converter transformers is progressively rising. Their reliability directly affects the safe and stable operation of both the equipment and the power grid. To study the degradation patterns of contact performance in electrical connectors, this paper establishes a wear degradation testing platform for electrical connectors in transformer oil, analyzing the evolution of contact performance and the characteristics of metallic microparticles under fretting wear. The study investigates the mechanism by which adhesive wear affects the degradation of contact performance and proposes a failure mechanism for electrical connection components that integrates wear degradation. The research indicates that after 20,000 wear cycles, the silver-plated layer on the contactors is damaged, with significant contact surface damage and a fluctuation in contact resistance reaching 89.43μΩ, alongside the generation of numerous ellipsoidal, flake, and linear metallic debris, with size distributions ranging from a few micrometers to several hundred micrometers. Adhesive wear and abrasive wear are significant factors leading to the damage of the contactors plating and the degradation of the contact surface. Current and temperature are key accelerators of wear degradation. Under the cumulative effects of Joule heating, fretting wear, and chemical corrosion, contact performance gradually deteriorates, ultimately leading to contact failure. The findings of this study provide a theoretical basis for the analysis of overheating failures in electrical connection structures and the development and manufacturing of electrical connection components.

The structural optimization design of corrugated fins can improve the heat dissipation efficiency of distribution transformers. A high fidelity geometric model of a 10kV distribution transformer is constructed, it’s highly consistent with the geometric structure of the actual transformer. Then, the one-sixth model of the short side of the distribution transformer is cut, and the one-sixth model is simplified according to the analysis of the internal fluid flow. The influence of corrugated fin width W on the average oil temperature of detailed model and simplified model is compared, and the change pattern is basically consistent. Then, four key structural parameters, such as width W, length L, spacing J and thickness D, of the corrugated fins in the simplified model are constructed, with a group of W and L which have a great influence on the heat dissipation area and a group of D and J which have a great influence on the air flow. Through multi-parameter analysis, the influence pattern of the four parameters on the average oil temperature is obtained. Finally, the oil temperature change of the parameters W and L with great influence is analyzed, and the optimal design values are W=290 mm, L=600mm, D=8.4mm and J=45mm. Compared with the infrared data of the external maximum temperature before and after the optimization, the temperature rise after optimization decreased by 3.1°C, which can increase the life of transformer by 1.4 times.

Insulation coordination at the transformer terminals constitutes a critical step in the optimized design of converters. A two-dimensional axisymmetric parametric model with four-level refinement was developed to meet the requirements of multi-physics simulations, including electromagnetic and thermal fields. Based on a second-level simplified model, gradient electric potentials were applied as boundary conditions for both the valve-side and grid-side windings to simulate the electric field distribution. To optimize the field distribution at the winding ends, Monte Carlo sampling was employed to generate training data, and a computationally efficient surrogate model was constructed to replace time-consuming finite element simulations. The error between the surrogate model and the full simulation results remains below 1

This paper presents a trapezoidal toothed log-periodic ultra-high frequency antenna for partial discharge (PD) detection in high-voltage power equipment. The proposed antenna features a novel trapezoidal toothed structure and a systematic parametric optimization process, achieving an impedance bandwidth of 0.5–3 GHz (VSWR < 5) and a maximum gain of 6.34 dBi at 1.75 GHz. Experimental validation in a controlled sulfur hexafluoride (SF6) gas environment demonstrates the antenna’s ability to detect weak PD signals with a high signal-to-noise ratio of up to 20 dB and generate clear phase-resolved partial discharge (PRPD) patterns. The obtained PRPD patterns aligning well with theoretical expectations and commercial detection systems, highlighting the antenna’s potential for real-time PD monitoring and fault diagnosis in high-voltage power systems.

Epoxy-Al2O3 composites is the main material of HVDC GIL insulator, whose space charge characteristics are closely related to the safety of device. In production, semiconducting electrodes are respectively coated on the surface of the high voltage rod and grounding inserts to increase the bonding strength between the metal and the insulator. This paper studies the effect of semiconducting electrode on the space charge behavior of EA composites. In this paper, the EA composites for GIL insulator are firstly prepared. The epoxy-based semiconducting electrode (ER) and chloroprene rubber-based semiconducting electrode (CR) are coated on one side of the EA-1 and EA-2 samples, and then sputtering silver electrode (Ag). And Ag is the cathode and anode of EA-0. The space charge characteristics in each samples are measured by pulsed electroacoustic method under different polarity electric fields at 25 °C, 40 °C 60 °C and 80 °C. The results show that the number of electrons and holes injected from the ER and CR is similar, and the distribution of space charge seems to be less affected by the semiconducting electrode at low temperature. But the effect of ER and CR on charge distribution is obviously different at high temperature, and ER has better performance than CR in controlling charge injection.

Dielectric polymer capacitors are essential for electrostatic energy storage but suffer from charge transport‐induced energy losses, particularly at elevated temperatures where thermally activated charge carriers exacerbate conduction. Conventional mitigation strategies rely on introducing heterogeneous interfaces to create charge traps, complicating scalable film fabrication. A homogeneous molecular trapping mechanism would circumvent these complexities, yet remains underexplored. Herein, a charge trapping strategy is devised by modifying the lowest occupied molecular orbitals of dielectric polymers through Lewis acid‐base adduct formation. The use of tris(pentafluorophenyl)boron (BCF) as a Lewis acidic molecular additive introduces deeper charge traps in commercial polyetherimide (PEI) while retaining homogeneity. With only 0.5 wt.% loading, the PEI‐BCF film exhibits greatly improved breakdown strength, achieving an ultrahigh discharged energy density of 7.3 J cm −3 with excellent cycle stability at 200 °C. This work establishes a facile molecular approach to decoupling charge trapping from heterogeneous interfaces, enabling high‐energy‐density polymer capacitors operable under extreme thermal conditions.

Epoxy resin serves as a critical insulating component in ultra-high voltage dry direct current bushings. However, the accumulation of space charges within the epoxy resin, a byproduct of charge mobility, poses a significant risk to the reliability and operational safety of bushings. Conventional space charge attenuation models, especially after voltage removal, have limited use in simulations aimed at understanding this phenomenon. This study introduces an improved model integrating the bipolar charge transport mechanism with space charge decay based on the hopping conduction mechanism and Schottky's theorem, establishes a theoretical framework for predicting the space charge behavior following voltage removal, and conducts a simulation to investigate the decay process within the internal structure of epoxy resin and measure the residual charges after voltage removal using the pulsed electro-acoustic method. The experimental data validate the accuracy of the proposed model and theoretical assumptions. The findings show that the remaining negative charges after voltage removal are not enhanced by the field enhancement; the anodic positive and cathodic positive charges are distributed in two-segment discontinuous traps, whereas the negative charges are distributed in one-segment traps. The model identifies two distinct trapping sites for anodic and cathodic positive charges, and one trapping site for negative charges. After voltage removal, when the field strength exceeded 40 kV mm(-1), positive charges exist near the upper and lower electrodes, and negative charges exist near the center of the specimen. The consistency between the simulated predictions and experimental data proves the effectiveness of the proposed model in accurately simulating the space charge decay in epoxy materials after voltage removal.

The lumbar intervertebral disc exhibits a complex physiological structure with interactions between various segments, and its components are extremely complex. The material properties of different components in the lumbar intervertebral disc, especially the water content (undergoing dynamic change as influenced by age, degeneration, mechanical loading, and proteoglycan content) - critically determine its mechanical properties. When the lumbar intervertebral disc is under continuous pressure, water seeps out, and after the pressure is removed, water re-infiltrates. This dynamic fluid exchange process directly affects the mechanical properties of the lumbar intervertebral disc, while previous isotropic modeling methods have been unable to accurately reflect such solid-liquid phase behaviors. To explore the load-bearing mechanism of the lumbar intervertebral disc and establish a more realistic mechanical model of the lumbar intervertebral disc, this study developed a solid-liquid biphasic, fiber-reinforced finite element model. This model was used to simulate the four movements of the human lumbar spine in daily life, namely flexion, extension, axial rotation, and lateral bending. The fluid pressure, effective solid stress, and liquid pressure-bearing ratio of the annulus fibrosus and nucleus pulposus of different lumbar intervertebral discs were compared and analyzed under the movements. Under all the movements, the fluid pressure distribution was closer to the nucleus pulposus, while the effective solid stress distribution was more concentrated in the outer annulus fibrosus. In terms of fluid pressure, the maximum fluid pressure of the lumbar intervertebral disc during lateral bending was 1.95 MPa, significantly higher than the maximum fluid pressure under other movements. Meanwhile, the maximum effective solid stress of the lumbar intervertebral disc during flexion was 2.43 MPa, markedly higher than the maximum effective solid stress under other movements. Overall, the liquid pressure-bearing ratio under axial rotation was smaller than that under other movements. Based on the solid-liquid biphasic modeling method, this study more accurately revealed the dominant role of the liquid phase in the daily load-bearing process of the lumbar intervertebral disc and the solid-phase mechanical mechanism of the annulus fibrosus load-bearing, and more effectively predicted the solid-liquid phase co-load-bearing mechanism of the lumbar intervertebral disc in daily life.