In the presence of H ₂ O and O ₂ , the dissociation products of SF ₆ will decompose to form several main stable byproducts (i.e. SOF ₂ , SOF ₄ and SO ₂ F ₂ ) in an electrical discharge. These byproducts are chemically active and have been shown experimentally to be associated with discharge faults. However, the relationships between the discharges and types of decomposition components are still not clear, mainly due to the fact that the complex chemical processes during SF ₆ discharges are not fully understood. In order to comprehensively investigate the decomposition of SF ₆ , an approach combining density functional theory (DFT) and transition state theory (TST) was used to study the pathways of SF ₆ decomposition in mixtures with H ₂ O and O ₂ that involve electrically-neutral species. The complex chemical reactions were analyzed, and the corresponding rate constants were predicted. The structural optimizations, vibrational frequency calculations and zero-point energy calculations of the species involved in each chemical reaction considered were carried out using the DFT-B3LYP method. Single-point energies were calculated using the CCSD(T) method. Based on the energy information obtained, the rate constants were predicted by TST, over a large temperature range, from 300 to 12 000 K.

Partial discharge (PD) is not only an important symptom for monitoring the imperfections in the insulation system of a gas-insulated switchgear (GIS), but also the factor that accelerates the degradation. At present, monitoring ultra-high-frequency (UHF) signals induced by PDs is regarded as one of the most effective approaches for assessing the insulation severity and classifying the PDs. Therefore, in this paper, a deep learning-based PD classification algorithm is proposed and realized with a multi-column convolutional neural network (CNN) that incorporates UHF spectra of multiple resolutions. First, three subnetworks, as characterized by their specified designed temporal filters, frequency filters, and texture filters, are organized and then intergraded by a fully-connected neural network. Then, a long short-term memory (LSTM) network is utilized for fusing the embedded multi-sensor information. Furthermore, to alleviate the risk of overfitting, a transfer learning approach inspired by manifold learning is also present for model training. To demonstrate, 13 modes of defects considering both the defect types and their relative positions were well designed for a simulated GIS tank. A detailed analysis of the performance reveals the clear superiority of the proposed method, compared to18 typical baselines. Several advanced visualization techniques are also implemented to explore the possible qualitative interpretations of the learned features. Finally, a unified framework based on matrix projection is discussed to provide a possible explanation for the effectiveness of the architecture.

The rolling bearing is an important part of the train’s running gear, and its operating state determines the safety during the running of the train. Therefore, it is important to monitor and diagnose the health status of rolling bearings. A convolutional neural network is widely used in the field of fault diagnosis because it does not require feature extraction. Considering that the size of the network model is large and the requirements for monitoring equipment are high. This study proposes a novel bearing fault diagnosis method based on lightweight network ShuffleNet V2 with batch normalization and L2 regularization. In the experiment, the one-dimensional time-domain signal is converted into a two-dimensional Time-Frequency Graph (TFG) using a short-time Fourier transform, though the principle of graphics to enhance the TFG dataset. The model mainly consists of two units, one for extracting features and one for spatial down-sampling. The building units are repeatedly stacked to construct the whole model. By comparing the proposed method with the origin ShuffleNet V2, machine learning model and state-of-the-art fault diagnosis model, the generalization of the proposed method for bearing fault diagnosis is verified.

The gas insulated switchgear (GIS) plays a significant role in the power system. Partial discharge (PD) detection is an effective method for the operation security of GIS. This paper reviews numerous research works about PD source localization and proposes a novel method based on image edge detection and support vector machine (SVM). The energy density distribution of the ultra-high frequency signal is obtained by time-frequency analysis, then the edge features are extracted using Canny algorithm. The Canny algorithm is improved for adaptive threshold selection. The last distinguishable edge point coordinate of electromagnetic wave mode varies with the distance between PD source and detection point by reason of velocity dispersion characteristic. An equation is proposed to identify the longitudinal position of PD source, which is also verified in the experiment. SVM is trained to determine the circumferential positions of PD source. Finally, the procedure of the PD source localization method is summarized.

Due to the high global warming potential (GWP) and increasing environmental concerns, efforts on searching the alternative gases to SF₆, which is predominantly used as insulating and interrupting medium in high-voltage equipment, have become a hot topic in recent decades. Overcoming the drawbacks of the existing candidate gases, C5- perfluorinated ketone (C5 PFK) was reported as a promising gas with remarkable insulation capacity and the low GWP of approximately 1. Experimental measurements of the dielectric strength of this novel gas and its mixtures have been carried out, but the chemical decomposition pathways and products of C5 PFK during breakdown are still unknown, which are the essential factors in evaluating the electric strength of this gas in high-voltage equipment. Therefore, this paper is devoted to exploring all the possible decomposition pathways and species of C5 PFK by density functional theory (DFT). The structural optimizations, vibrational frequency calculations and energy calculations of the species involved in a considered pathway were carried out with DFT-(U)B3LYP/6-311G(d,p) method. Detailed potential energy surface was then investigated thoroughly by the same method. Lastly, six decomposition pathways of C5 PFK decomposition involving fission reactions and the reactions with a transition states were obtained. Important intermediate products were also determined. Among all the pathways studied, the favorable decomposition reactions of C5 PFK were found, involving C-C bond ruptures producing Ia and Ib in pathway I, followed by subsequent C-C bond ruptures and internal F atom transfers in the decomposition of Ia and Ib presented in pathways II + III and IV + V, respectively. Possible routes were pointed out in pathway III and lead to the decomposition of IIa, which is the main intermediate product found in pathway II of Ia decomposition. We also investigated the decomposition of Ib, which can undergo unimolecular reactions to give the formation of IV a, IV b and products of CF₃ + CF-CF₃ in pathway IV. Although IV a is dominant to a lesser extent due to its relative high energy barrier, its complicated decomposition pathway V was also studied and CF₃, C = CF₂ as well as C-CF₃ species were found as the ultimate products. To complete the decomposition of C5 PFK, pathway V I of Ic decomposition was fully explored and the final products were obtained. Therefore, the integrate decomposition scheme of C5 PFK was proposed, which contains six pathways and forty-eight species (including all the reactants, products and transition states). This work is hopeful to lay a theoretical basis for the insulating properties of C5 PFK.

The ultra-high frequency (UHF) method is widely used in insulation condition assessment. However, UHF signal processing algorithms are complicated and the size of the result is large, which hinders extracting features and recognizing partial discharge (PD) patterns. This article investigated the chromatic methodology that is novel in PD detection. The principle of chromatic methodologies in color science are introduced. The chromatic processing represents UHF signals sparsely. The UHF signals obtained from PD experiments were processed using chromatic methodology and characterized by three parameters in chromatic space (H, L, and S representing dominant wavelength, signal strength, and saturation, respectively). The features of the UHF signals were studied hierarchically. The results showed that the chromatic parameters were consistent with conventional frequency domain parameters. The global chromatic parameters can be used to distinguish UHF signals acquired by different sensors, and they reveal the propagation properties of the UHF signal in the L-shaped gas-insulated switchgear (GIS). Finally, typical PD defect patterns had been recognized by using novel chromatic parameters in an actual GIS tank and good performance of recognition was achieved.