Abstracts:Furfural is a chemical marker that characterizes the aging state of paper insulation in power transformers. Current research lacks reports on the prediction of furfural concentration in insulating oils, which is not favorable for predicting the potential failure or aging state of transformer paper insulation. In light of this, this article presents a method for predicting furfural concentrations utilizing the multistrategy particle swarm optimization (MSPSO) combined with the back propagation neural network (BPNN) algorithm (MSPSO-BPNN), and thus evaluates the aging state of the insulating paper. First, oil-paper insulation samples are prepared under laboratory conditions, and concentrations of furfural in oil are measured at different aging periods; second, the furfural concentration time-series data are predicted using the MSPSO-BPNN algorithm, and the root mean squared error (RMSE) of the predicted results reached 0.05; ultimately, based on the predicted furfural concentration data and the furfural-polymerization degree (DP) assessment model considering the second-order aging kinetic theory, the aging assessment of the insulating paper has been achieved with an accuracy of 98.7%. Therefore, this work can provide technical and theoretical support in the assessment of the aging state of insulating paper through the application of furfural concentration prediction.

Abstracts:The corrosion and ablation of the water-blocking buffer layer of high-voltage cables are accompanied by the production of various gases. By selecting characteristic gases and analyzing the corresponding relationship between their concentration and the corrosion/ablation stages, it is possible to evaluate the status of the buffer layer. This article analyzes the time-varying characteristics of gas concentration during the corrosion and ablation of water-blocking buffer layers, reveals the mechanisms of gas production, and proposes a method for evaluating the status of the buffer layer based on characteristic gas analysis. Results indicate that both the corrosion and ablation processes of the buffer layer produce H2, CH4, and CO2, but the ablation process generates additional gases including C2H6, C2H4, C2H2, and CO. The H2 concentration is positively correlated with the amplitude of applied voltage, water content, and surface pressure and saturates after a period of generation. On this basis, the mechanism of H2 production during the corrosion stage and the decomposition of polyethylene terephthalate (PET) accompanied by hydrocarbon gas production during the ablation stage of the buffer layer are investigated. By detecting the concentration of H2 in the cable, the overall damage level of the buffer layer can be determined. Additional detection on C2H6, C2H4, C2H2, and CO gases can determine whether the buffer layer is in a corrosion state. Finally, the effectiveness of evaluating the status of the buffer layer by characteristic gas analysis is proved on new and on-site 110-kV cables.

Abstracts:The leading position among liquid dielectrics used in high-voltage equipment is occupied by transformer oil. Over 130 years of using insulating oil, a lot of valuable information has been obtained, in particular: about the condition of the oil during its operation; about methods for restoring its characteristics; known methods of protection against active oxidation; and methods for monitoring oil parameters and assessing the condition of the insulation system of oil-filled equipment have been developed. Despite the fact that mineral oils cope quite well with their functions—insulation and heat removal, there are a number of properties that do not meet the requirements of modern energy. Thus, in terms of fire resistance, chemical stability, and biodegradability, transformer oils are significantly inferior to ester dielectric fluids, which are quite widely used by foreign energy companies. In addition, transformer oils, during their prolonged use in electrical equipment, are capable of forming sediments of various consistencies. For efficient and reliable operation of the insulation system, such precipitation is extremely undesirable. This article proposes a method for optimizing the properties of transformer oil (aromatic and paraffin) to stabilize the sedimentation process by mixing the oil with a synthetic ester in a certain ratio.

Abstracts:Partial discharge pattern recognition (PDPR) is the fundamental cornerstone for fault diagnosis. It has emerged as a pivotal focal point in the field of power systems. However, PDPR faces several challenges, such as low signal quality and complex discharge patterns. We propose a multiscale residual aggregation transformer network (MRATNet) to address these challenges effectively. MRATNet learns long-dependent semantic relationships and discriminative features in partial discharge (PD) signals. Moreover, it integrates convolutional and transformer architectures as the feature extraction backbone. Thus, multiscale residual convolution (MSRC) blocks are incorporated to aggregate diverse information, and the transformer is leveraged to capture long-dependent semantic relationships. Meanwhile, the cross-attention mechanism is introduced to capture the spatial and channel feature distributions. The composite embedded feature selection (CEFS) module is proposed to extract discriminative features. Comprehensive experiments demonstrate the effectiveness of MRATNet, yielding exceptional performance on DEPD dataset (91.47%) and PDMDB dataset (86.05%). Finally, extended experiments have been conducted using the Technical University of Berlin’s German emotional language library, suggesting the potential for generalizing our method to other recognition tasks.

Abstracts:The wind-turbine step-up transformers are subjected to high-frequency high dV/dt transient voltages due to the operation of power electronic converters and vacuum circuit breakers. These high-frequency transients are harmful to the transformer turn-to-turn and layer-to-layer paper-oil insulation. The current research work addresses the effect of the transient voltage parameters on the degradation of the transformer paper-oil insulation. Two-level, two-factor design of experiments were used to obtain the effect of transient voltage parameters on the aging of the paper-oil insulation. Rise time and switching frequency were the two factors (parameters) selected for the aging experiments. The two values of rise time were 220 and 650 ns, and two values of the switching frequency were 1 and 3 kHz. The faster rise time (220 ns) and higher switching frequency (3 kHz) of the voltage led to a higher degradation for the paper-oil insulation. In addition, the transient voltage aging of paper-oil insulation is investigated and compared for two types of mineral oils: Luminol TRi oil and Voltesso 35 oil, which have differences in their electrical and chemical properties.

Abstracts:Corrosive sulfur compounds in mineral insulating oils can easily react with metal surfaces in power transformers (PTs) and form deposits, which can further cause failures of the electrical equipment. Copper sulfide formation is well studied over the last couple of decades. However, silver sulfide formation also requires attention, due to a significant number of failures of on-load tap changers (OLTCs) caused by silver corrosion reported worldwide from different utilities. The conventional silver corrosion tests (DIN 51353 and ASTM D1275-15) do not provide an adequate assessment of oil corrosivity against silver, even if the oil contains dibenzyl disulfide (DBDS) and other reactive sulfur compounds (disulfides and mercaptans) in high concentrations. The new testing method for silver corrosion is performed on a significant portion of oils containing DBDS, as a major reactive sulfur compound, which remained “noncorrosive” following the ASTM D1275-15 silver strip test. The method is demonstrated on five oils from free-breathing PTs in service, containing DBDS in a wide range of concentrations (from approximately 40 to 200 mg/kg of DBDS). The new method comprises a modification to the ASTM D 1275–15 test setup, by introducing longer test duration times and continuous oxygen ingress, which is found to be the most crucial parameter for promoting the formation of silver sulfide deposits. This is confirmed by energy-dispersive X-ray (EDX) results, showing an increased sulfur weight percentage on the silver strip after the test. The application of the modified ASTM silver corrosion test will contribute to improving the assessment of sulfur corrosion risk.

Abstracts:In light ice-covered areas, icing of composite insulators with different pollution degrees has a great influence on their insulation performance. In this article, the solid layer pollution method was adopted to realize different degrees of pre-pollution for six kinds of FXBW-220/160 composite insulators. Then, with the reference ice thickness of 10 mm as the standard, icing and ac flashover tests were carried out on insulators under the multifunctional artificial climate chamber. The ice weight and minimum flashover voltage ( ${U}_{\textit {mf}}{)}$ of insulators under different pollution degrees were obtained, and the path of the flashover arc was recorded. In addition, the maximum withstand voltage gradient equation of insulators under different discharge probabilities was established. The results show that when insulators are energized icing, the ice weight decreases with the increase of salt deposit density (SDD), which in turn causes the ${U}_{\textit {mf}}$ to decrease. The relationship between the ${U}_{\textit {mf}}$ and SDD is a negative power function regardless of whether the insulators are energized icing or not. The differences in arc path, residual ice layer resistance, and the ratio of air gap arc to ice surface arc are the main reasons for affecting the flashover voltage. In addition, the maximum withstand voltage gradient ${E}_{\alpha }$ at discharge probability $\alpha $ % of 10%, 5%, and 0.13% when the insulators are energized icing decreases with the increase of SDD, and ${E}_{{10}} > {E}_{{5}} > {E}_{{0}.{13}}$ . Increasing the spacing and diameter of the largest shed of composite insulators and reasonably cooperating with medium/small sheds can help to significantly improve their flashover performance. The results of this article can provide data support for the selection and optimal design of insulators for 220 kV transmission lines in light ice-covered areas.

Abstracts:Suppressing the activity of metal particles is a key issue in the development of dc gas-insulated power transmission lines (GILs). This study investigates the effects of five different dielectric coatings of the electrodes—namely surface oxidation treatment and four different coating materials (i.e., 1032K high-resistance paint (1032K), teflon, epoxy marlin orange primer (EM primer), and iron red primer (IR primer) spray)—on the movement characteristics of ball and block metal particles under dc voltage and polarity reversal voltage in an SF6 gas environment at an absolute pressure of 0.3 MPa. The results showed that the surface oxidation treatment of the electrodes had little effect on suppressing the particle activity. The coating in the high-voltage electrode did not significantly change the particle jumping voltage, and once the particles jumped, they were adsorbed onto the coating. Coating the ground electrode significantly increased the jumping voltage of the particles. Under dc voltage, the lifting field strength of the particles exceeds 5 kV/mm which is much higher than the maximum design field strength at the ground electrode of dc GIL. The increase in coating thickness did not significantly improve the jumping field strength. Under polarity reversal voltage, the jumping field strength of the particles was lower than under dc. However, the coating of the ground electrode also had a significant effect on suppressing the particles during the reversal process. Simulation results showed that the coating hindered the charge exchange between the particles and the electrode, increased the field strength at the contact position between the particles and the electrode, and reduced the field strength on the side of the particles away from the electrode.

Abstracts:This article constitutes a brief account of an extensive investigation into different stages of first-stroke negative lightning leader stepping, starting with the determination of the length of the negative streamer zone below the leader tip. An already formed space stem/space leader results in the propagation of a positive space leader, which bridges the streamer zone. This takes a time too short for leader thermalization. Such thermalization occurs over a much longer time, through corona current injected at the tip of the moving negative space leader. The latter stage is decisive for the time interval between steps. The model has then been applied to leaders with prospective return stroke current of 3–100 kA, leader tip height of 23–3000 m, and leader charge linearly decaying to practically vanish at heights of either 2500 or 4500 m. The dependence of the negative leader stepping characteristics on the above parameters is shown in the following. Under the conditions investigated, the step length varied between 3.5 and 39 m, the bridging time of the streamer zone in the range of 0.8– $3~\mu \text{s}$ , the time interval between steps in the range of 8.5– $92~\mu \text{s}$ corresponding to a mean stepped leader speed in the range of 1.0– $11.6^{\ast} 10^{{5}}$ m/s. Whenever possible, the model predictions were satisfactorily compared to field measurements. Triggered lightning, dart leaders, and the attachment process to ground objects are out of the scope of this investigation.

Abstracts:Retrofilling transformers with ester fluids have been increasingly practiced. Transformers may be retrofilled at different service years. In this article, the aging behavior of the cellulose insulation of transformers retrofilled at different service years (with different conditions of paper insulation) is studied. Since the temperature may rise after the retrofill due to the high viscosity of the ester liquids, a temperature increase of 10 °C and 20 °C in cellulose-liquid insulation systems after retrofill is considered. The degree of polymerization (DP) of the cellulose insulation and the moisture in both cellulose and liquid insulation are measured before and after retrofilling. The measurements are used for modeling the remaining useful life of the cellulose-liquid insulation system. The process of insulation systems reaching the impregnation equilibrium after retrofilling is investigated through the fire and flash point measurements of the liquid insulation and frequency domain spectroscopy (FDS) measurements of the cellulose-liquid insulation system.