As offshore wind power expands into deep sea, the precise calculation of steady-state ampacity for high-voltage direct current (HVDC) submarine cable is of critical importance. Existing studies often neglect the influences of ocean currents on the heat transfer in seabed porous media and submarine cable laying conditions, leading to deviation in ampacity assessment. This study established a three-dimensional thermo-electro-hydrodynamic coupled model based on COMSOL multiphysics to systematically investigate the influencing mechanisms of sediment permeability, porosities, and submarine cable laying conditions on the temperature field and ampacity of HVDC XLPE submarine cable under vertical ocean currents. The results show that seawater flow velocity is the decisive factor influencing ampacity. Setting the sediment porosities to 0.4 and the permeability to 10³ d, when the flow velocity increases from 2 cm/s to 50 cm/s, the ampacity increases by 35.2%, which is mainly attributed to the significant enhancement of convective heat transfer efficiency in the sediment layer. Setting the sediment permeability to 10³ d and the seawater flow velocity to 50 cm/s, when the sediment porosities increases from 0.2 to 0.6, the ampacity increases by 5.6%, which attributes to expanded pore networks enhancing the equivalent thermal conductivity. Sediment permeability and flow velocity exhibit a significant synergistic effect, when the permeability exceeds 10² d, secondary convective heat transfer can be induced in high-speed flow fields. Setting the sediment porosities to 0.4 and seawater flow velocity to 50 cm/s, compared to 10² d of sediment permeability, the flowing fluid can additionally contribute 1.6% of the ampacity gain when the sediment permeability is 10³ d. The bipolar system exhibits a significant thermal coupling effect under ocean currents, causing the temperature difference between the two poles to increase with the increase of flow velocity. Setting the sediment porosities to 0.6 and the permeability to 10³ d, when the flow velocity is 50 cm/s, the temperature difference between the two poles increases by 1 283.3%, resulting in an ampacity difference up to 136 A. Among the laying parameters, the laying spacing has a greater impact on the ampacity than the laying depth.

Regarding the issue of metal suspended particle defects in basin insulators, based on the electroluminescence principles, a ZnS:Cu/epoxy resin luminescent composite coating was prepared by doping sulfurized zinc-based material (ZnS:Cu) into epoxy resin (EP), and its properties were characterized. The luminescence pattern of the coating was investigated using rod-plate electrodes, and a defect experiment was simulated on basin-type insulator to explore the feasibility of the ZnS:Cu/epoxy resin coating in defect self-diagnosis. The results show that with the increase of ZnS:Cu content, the surface flashover voltage of the ZnS:Cu/epoxy resin composite coating rises, and the dielectric loss factor increases. The composite coating has a certain luminous threshold, when the mass fraction of ZnS:Cu is 10%-30% and the coating thickness is 0.1-0.3 mm, its luminous intensity increases with the increase of electric field strength, ZnS:Cu content, and coating thickness. Moreover, after spraying the ZnS:Cu/epoxy resin coating on the basin insulator, the defect location can be determined through the luminescence, achieving the purpose of defect diagnosis, which provides a new idea for the defect identification and diagnosis of power insulation equipment.

In order to investigate the effects of degassing treatment on the electrical tree growth and discharge evolution in DC cable insulation, and to clarify the affecting mechanism of degassing treatment in the critical breakdown process of electrical tree, taking cross-linked polyethylene (XLPE) samples with different degassing time as research objects, on the basis of partial discharge (PD) experiment data, the dynamic evolution processes of electrical tree growth rate and discharge magnitude were quantified by combining derivative analysis and adaptive threshold method, and the critical breakdown characteristics were identified by counting mutation point. The results show that degassing treatment can significantly delay electrical tree growth, reduce discharge activity, and increase the breakdown threshold. In short-time degassed samples, charges tend to accumulate more rapidly and trigger electrical tree expansion immediately, and the electrical tree growth and the discharge evolution are synchronous. However, when degassing is sufficient, the processes of trap capturing charges and energy accumulation become more prominent, and the discharge shows significant lag in driving electrical tree propagation, and drives the rapid evolution close to breakdown through phased accumulation–release mode. The proposed analytical method can effectively improve the identification accuracy of breakdown point, providing a theoretical basis for the insulation optimization and health monitoring of DC cables.

First, the dimensions and material properties of imported hot sticks were measured and characterized to determine their mechanical and electrical performance parameters. Then, based on the parameters of the imported products and the existing manufacturing process of hot sticks, simulation design and process optimization were conducted. Finally, a hot stick was fabricated using pultrusion process, and its mechanical and electrical performance were tested. The results show that on the basis of mechanical and electrical physical fields simulations and systematic experimental validation, a domestic lightweight hot stick is successfully developed, and its maximum deflection difference does not exceed 13.2 mm. After 9 times of withstand voltage test under dry conditions, no breakdown, flashover or overheating occur in the hot stick. Moreover, compared with the ordinary domestic hot stick, the prepared hot stick has reduced its unit weight by 29.9%.

Aiming at the problem that the transformer life prediction model has low fit and ignores the influence of the operating state, a residual life prediction model (POA-RF model) based on pelican algorithm (POA) optimized random forest (RF) model was proposed. Taking preventive tests, online monitoring, defect reports, and account data as research objects, a transformer health index system consisting of transformer body, oil quality, and dissolved gas in oil health indices was constructed. Considering the corrective effect of operating status on residual life, with relevant parameters of the health index system as inputs, the POA was used to optimize the RF model to solve its overfitting problem and improve prediction accuracy. Experiments are conducted with the collected transformer instance data to verify the accuracy of the model. The results show that for the 8 groups of scrapped samples in the test samples, the average absolute error of the residual life predicted by the POA-RF model is 1.187 0 years, which is 20.52% and 49.02% lower than the predicted value of the RF model and the calculated value of the health index theory, respectively. For all the test samples, compared with other commonly used machine learning algorithms, the relevant performance indicators of the POA-RF model are all the best, and the optimization effect is obvious, which can better predict the residual life of transformer.

In order to investigate the performance degradation characteristics of multi-layer Nomex insulating paper used in transformers during the thermal ageing process, firstly, the multi-layer oil-immersed Nomex insulating paper was conducted accelerated thermal ageing test, then a frequency domain dielectric spectroscopy testing platform and a surface potential testing platform were established to obtain the dielectric loss factor and surface potential decay curves of oil-immersed insulating paper at different ageing time, and the surface trap distribution characteristics of the oil-immersed insulating paper with different ageing degree were calculated and analyzed. The results show that with the increase of thermal ageing degree, the dielectric loss factor of oil-immersed Nomex insulating paper samples increases significantly across all frequency bands. The dielectric loss factors of double-layer and triple-layer oil-immersed insulating paper samples are significantly higher than that of single-layer oil-immersed insulating paper sample, and the dielectric loss factor of triple-layer oil-immersed insulating paper sample exhibits an even greater increase. As the thermal ageing degree increases, the initial surface potentials of the insulating paper samples gradually decrease, and the surface potential decay rate increases with ageing time, while at the same ageing cycle, the surface potential decay rate gradually decreases with the increase of insulating paper layer number. The shallow trap density of oil-immersed insulating paper increases with the increase of ageing degree, while the deep trap density decreases. The decline rates of deep trap density and energy level of double-layer oil-immersed insulating paper are lower than those of single-layer oil-immersed insulating paper, while the increase rates of its shallow trap density and energy level are higher than those of single-layer oil-immersed insulating paper. The research results can provide some reference for the design of transformer oil-paper insulation.

In this paper, a novel crosslinked network cellulose insulating paper was prepared by citric acid crosslinking modification, and the dielectric constant and dielectric loss factor of the insulating paper were measured before and after modification. Then the composite insulation system of the modified insulating paper and mineral insulating oil were carried out partial discharge and electrical life tests. The results show that compared with the unmodified insulating paper, the dielectric constant and dielectric loss factor of the novel crosslinked network insulating paper decreases by 15.39% and 29.78%, respectively, and the electrical life index increases by 37.94%. The novel crosslinked network insulating paper-insulating oil composite insulation system has lower discharge repetition rate, discharge energy, average discharge, and maximum discharge.

In order to obtain the characteristic gases generated during the decomposition of cross-linked polyethylene (XLPE) at different stages under electrical and thermal factors to judge the degree of degradation, we systematically studied the gas production characteristics of XLPE under thermal field and electrical field through thermogravimetric-infrared (TG-IR) experiments, high-energy discharge gas production experiments, and molecular dynamics (MD) simulations based on reactive force field (ReaxFF). The results show that under the action of thermal field, XLPE primarily decomposes to generate ethylene, propylene, and long-chain hydrocarbons, accompanied by oxidation products such as CO and CO₂. While under the action of electrical field, the characteristic gases are mainly acetylene and propyne, and distinct absorption peaks of propyne appear in the infrared spectrum under high-energy discharge. Molecular simulation further reveals the differences in decomposition pathways of XLPE under electric and thermal fields. Specifically, decomposition under the thermal field is dominated by end-group breakage, while the electric field accelerates free radical reactions and promotes the generation of highly unsaturated substances. Finally, a normalized characteristic gas infrared spectrum library is constructed through density functional theory (DFT) calculation. The gas production process is divided into different stages based on the turning point of thermogravimetric curve, the peak spectrum change temperature of gas infrared spectrum, and the molecular simulation trajectory time of the decomposition process. This study clarifies the characteristic gases of XLPE during electrical/thermal decomposition and the gas production mechanisms, which can provide a theoretical basis for the fault detection of XLPE cable based on gas infrared spectrum.

Cross-linked polyethylene (XLPE) submarine cable is a key transmission equipment in flexible low-frequency transmission scenarios such as medium and long-distance island transmission and offshore wind-solar new energy power generation. In this paper, XLPE insulated cable was conducted electrothermal ageing tests under low frequency voltage, and the effects of ageing voltage frequency on the electrical properties, mechanical properties, and micro-structure of XLPE insulation were studied. The results show that when the ageing voltage frequency is 10-50 Hz, the AC electric strength, volume resistivity, and elongation at break of XLPE ageing samples increase at first and then decrease with the increase of ageing voltage frequency. The electrical properties of XLPE samples aged under 20 Hz and 30 Hz of voltages are better than those aged under other voltage frequencies. The change trend of crystallinity of XLPE ageing samples with ageing voltage frequency is consistent with the electrical properties. The thermal ageing at lower voltage frequency contributes to the recrystallization of XLPE insulating materials, while the thermal ageing at higher voltage frequency will damage the crystalline structure of the material and increase the proportion of amorphous regions. The research results can provide a reference for the applicability of XLPE insulated cables in low frequency transmission technology.

Degassing heat treatment is the longest process in the production of cross-linked polyethylene (XLPE) cables, and reducing degassing time is an important way to improve cable production efficiency. Focused on the migration of residual by-products in XLPE insulating materials for high-voltage cables, we established a diffusion model of cross-linked by-products driven by concentration gradient. The concentration of by-products in the insulation of a 320 kV DC cable after degassing was measured and compared with the simulation results, and the effects of temperature and initial concentration of by-products on the degassing process were investigated. The results show that the simulation and measured results of the concentration distributions of methane and acetophenone in the 320 kV DC cable after degassing are essentially consistent, verifying the correctness of the model. Higher temperatures can accelerate the degassing process of by-products. Within the temperature range of 65-75℃, for every 5℃ increase in temperature, the degassing time for methane is shorten by approximately 30%, and the degassing time for acetophenone is shorten by approximately 72 h. Reducing the initial by-product concentration can also shorten the cable degassing time. It is an important way to reasonably regulate the initial concentration of by-products and degassing temperature for optimizing the cable degassing process and improving enterprise efficiency.