Surface charges have significant impact on the evolution of surface dielectric barrier discharge (SDBD). In this work, the role of residual surface charges on repetitively nanosecond pulsed SDBD in atmospheric air is investigated using a two-dimensional fluid model, based on the assumption of preserving the distribution of surface charges at the end of the previous high voltage (HV) pulse. In the bipolar mode when the polarity of residual surface charges is opposite to that of the current HV pulse, a lower breakdown voltage and more deposited energy can be observed, showing an obvious enhancement of SDBD. In the unipolar mode, residual surface charges suppress the development of discharges and energy deposition. It is found that more residual surface charges are accumulated during the negative pulsed discharge, which have a more pronounced effect on the subsequent positive pulsed one. This is explained by the fact that the negative surface streamers directly contact the dielectric and charge it, while the positive surface streamers float above the dielectric, forming a ion-rich region near the surface. The results in this work demonstrate the mechanism of how residual surface charges affect discharge dynamics, which can be utilized to regulate energy deposition in SDBDs.

Oil chromatography data together with its variation trend provides the key basis for the evaluation of the transformer health state. Existing studies on Deep Belief Network (DBN), support vector machine (SVM), et al. have reported a few results in the field of oil chromatography data prediction. However, the above methods are proposed for the classification problem, so there is no theoretical basis for applying the above methods to the time-series prediction. The wrong usage limits the accuracy of the predicted results, which is observed as the obvious “time-shift” error in the prediction curve, leading to the predicted state inconsistent with the actual situation. To fill the gap, a Deep Recurrent Belief Network (DRBN) model for transformer state prediction was proposed based on time series theory and oil chromatography data characteristics. In this model, the construction shortage of DBN in time series prediction was analyzed as a prototype pattern to structure the stereoscopic mapping relation. The self-adaptive delay network based on the principle of autocorrelation, together with the corresponding error feedback network, realized the stereoscopic flow of data in multi-dimensional space and thus ensured the model’s valid expression of time domain correlation. In addition, the cross-entropy loss function, based on Kullback-Leibler divergence and Weibull distribution with an obvious characteristic of DGA, was constructed to eliminate the uncertainty of initialization process and also effectively control the direction and step size of the error gradient. Examples in the field were used to verify the method to find that the model proposed in this paper can availably overcome the “time-shift” error, and its prediction accuracy can reach more than 95.16%.

Nanoparticles currently in use are challenged in further improving the dielectric strength of insulating oil. There is a great need for a new type of nanoparticle to promote the application of insulating oil-based nanofluids in electric industries. This paper experimentally investigates the effect of nanoparticle morphology on pre-breakdown and breakdown properties of insulating oil-based nanofluids. The positive impulse breakdown voltage of insulating oil can be significantly increased by up to 55.5% by the presence of TiO₂ nanorods, up to 1.23 times that of TiO₂ nanospheres. Pre-breakdown streamer propagation characteristics reveal that streamer discharge channels turn into a bush-like shape with much denser and shorter branches in the nanofluid with TiO₂ nanorods. Moreover, the propagation velocity of streamers is dramatically decreased to 34.7% of that in the insulating oil. The greater improvement of nanorods on the breakdown property can be attributed to the lower distortion of the electric field. Thus, when compared with nanospheres, pre-breakdown streamer propagation of nanofluid is much more suppressed with the addition of nanorods, resulting in a greater breakdown voltage.

The flashover phenomenon of the insulator is the main cause for insulating failure of GIS/GIL, and one of the most critical impacting factors is the accumulation of surface charge. The common methods to restrain the surface charge accumulation are reviewed in this paper. Through the reasonable comparison and analysis of these methods, nano-coatings for the insulator were selected as a way to restrain the surface charge accumulation. Based on this, six nano-coated epoxy resin samples with different concentrations of P25-TiO ₂ nanoparticles were produced. A high precision 3D surface charge measurement system was developed in this paper with a spatial resolution of 4.0 mm ² and a charge resolution of 0.01 µ C (m ² · mV) ⁻¹ . The experimental results for the epoxy resin sample showed that with the concentration of nanoparticles of the coating material increasing, the surface charge density tended to first decrease and then increase. In the sample coated with 0.5% concentration of nanoparticles, the suppression effect is the optimum, leading to a 63.8% reduction of charge density under DC voltage. The application test for actual nano-coated GIS/GIL basin insulator indicated that the maximum suppression degree for the charge density under DC voltage could reach 48.3%, while it could reach 22.2% for switching impulse voltage and 12.5% for AC context. The control mechanism of nano-coatings on charge accumulation was proposed based on the analysis for surface morphology features and traps characteristics; the shallow traps dominate in the migration of charges while the deep traps operate on the charge accumulation. With the concentration of nanoparticles in nano-coating material mounting up, the density of shallow traps continuously increases, while for deep traps, it first decreases and then increases. For the sample with 0.5% concentration of nanoparticles coated, the competition between shallow traps and deep traps comes to the most balanced state, producing the most significant suppression impact on surface charge accumulation.

The healthy state of insulation in oil-impregnated bushings is traditionally evaluated by tanδ and capacitance at power frequency and mostly at 10 kV in the test standard. However, there has frequently been insulation accidents induced by moisture ingress (MI) for bushings that have passed the standard. The mechanism and new diagnostic features for MI into bushings were not distinct enough and an accurate test method is urgently needed research. To address this technical gap, a bushing model with a transparent sheath was designed and an ultrasonic humidifier device was adopted to simulate the environment of MI in bushings and recorded by digital camera. The parameters of dielectric dissipation factor, capacitance, partial discharge (PD), frequency domain response, and moisture content in oil were measured at room temperature with time. The results presented that both the increment dissipation factor at low frequency of 0.001 Hz and the increment dissipation factor of 1.2 U_m could be used for detecting the earlier insulation defect of oil-impregnated paper (OIP) bushings. The phase resolved partial discharge (PRPD) can serve as the diagnostic basis of the severe state (S₃) of insulation deterioration caused by MI into bushings around the phases of 0–117°, 151–303°, and 325–360°. The research findings would provide a useful reference for the condition diagnosis and maintenance of OIP bushings. Especially, the increment detection of Frequency Domain Spectroscopy (FDS) at the frequency of 1 mHz and 10 kHz was recommended firstly for the operative bushings in real sites.