Currently, the cleaning process for power equipment monitoring data is cumbersome and often leads to loss of information. To address these problems, a data cleaning method based on stacked denoising autoencoder (SDAE) networks is proposed in this paper. SDAE networks have a strong ability to denoise and restore corrupted data and have a strong feature extraction capability. The status monitoring data of equipment under normal conditions are trained by SDAE to obtain the cleaning parameters and the reconstruction errors. An upper threshold of the reconstruction errors obtained from training samples is determined through Kernel density estimation. A tolerance window width is added to achieve rapid anomaly detection. The abnormal data are classified as outliers, missing data, or fault status data according to the relationship between the reconstruction error and the threshold and between the duration of abnormal data and the tolerance window. To verify the effectiveness of the proposed method, the SDAE model is used to process the data for the dissolved gas concentration in transformer oil and the temperature of the transmission line. The results show that the proposed method can effectively identify and repair outliers and missing information. The model can perform rapid anomaly detection when the equipment is running abnormally.

The state of transformer equipment is usually manifested through a variety of information. The characteristic information will change with different types of equipment defects/faults, location, severity, and other factors. For transformer operating state prediction and fault warning, the key influencing factors of the transformer panorama information are analyzed. The degree of relative deterioration is used to characterize the deterioration of the transformer state. The membership relationship between the relative deterioration degree of each indicator and the transformer state is obtained through fuzzy processing. Through the long short-term memory (LSTM) network, the evolution of the transformer status is extracted, and a data-driven state prediction model is constructed to realize preliminary warning of a potential fault of the equipment. Through the LSTM network, the quantitative index and qualitative index are organically combined in order to perceive the corresponding relationship between the characteristic parameters and the operating state of the transformer. The results of different time-scale prediction cases show that the proposed method can effectively predict the operation status of power transformers and accurately reflect their status.

The severity evaluation of UHF signals of partial discharge in GIS is helpful to formulate a maintenance strategy in time, and improve the reliability of power system operation. Most of the evaluation methods are based on structured data, which cannot fully characterize the state of the equipment. In this paper, a method for evaluating the severity of UHF signals of partial discharge in GIS based on semantic analysis is presented. It comprehensively considers the influence of structured data and unstructured text data on the state of equipment, including measured data on partial discharge, relevant information of defects, and equipment operating parameters. Firstly, the severity of UHF signals of partial discharge is defined. According to the on-site detection of substations, a data set containing equipment detection reports and PRPS data is established. Aiming at semantic analysis, Word Embedding is used to associate and encode textual information to reduce the subjectivity of the encoding. Measured data on partial discharge is also combined and analyzed for evaluation. The method proposed is used for case analysis and compared with other methods. The results show that the comprehensive utilization of structured data and unstructured text data can reflect the state of equipment more comprehensively and truthfully, and improve the accuracy of the results.

With the accumulation of partial discharge (PD) detection data from substation, case-based reasoning (CBR), which computes the match degree between detected PD data and historical case data provides new ideas for the interpretation and evaluation of partial discharge data. Aiming at the problem of partial discharge data matching, this paper proposes a data matching method based on a variational autoencoder (VAE). A VAE network model for partial discharge data is constructed to extract the deep eigenvalues. Cosine distance is then used to calculate the match degree between different partial discharge data. To verify the advantages of the proposed method, a partial discharge dataset was established through a partial discharge experiment and live detections on substation site. The proposed method was compared with other feature extraction methods and matching methods including statistical features, deep belief networks (DBN), deep convolutional neural networks (CNN), Euclidean distances, and correlation coefficients. The experimental results show that the cosine distance match degree based on the VAE feature vector can effectively detect similar partial discharge data compared with other data matching methods.

Recently, the intelligent construction of distribution station is still in its infancy. Information is not fused between different monitoring systems. The current investment in distribution network detection means is large, which does not match the low-cost characteristics of distribution network equipment. So information isolation and high cost are the two main restrains of condition monitoring. Aiming at solving these problems, this paper proposes a new monitoring system for power distribution station in the context of the ubiquitous power Internet of things. Firstly, the overall design of the system is studied. A highly integrated comprehensive monitoring terminal is developed, based on the idea of integrated perception. It can reduce monitoring and communication costs through the flexible access and efficient processing of key state parameters such as sound, light, electricity, and heat. Secondly, algorithms are developed for evaluation and diagnosis of power distribution equipment. Artificial intelligence and big data analysis methods were used for condition evaluation and anomaly detection. It improves the efficiency of fault handling. Finally, the monitoring systems are applied in some power distribution stations. After a period of application, the system cross-professional data are fully integrated and multiple state parameters are obtained in real time for comprehensive application. In conclusion, this work can provide new solutions for the construction of intelligent power distribution stations.

Corona discharge theory has been widely applied in various industrial fields, including power transmission and transformation project, gas purification, and particle charging. However, its microphysical mechanism remains a hot research topic. Currently, the theoretical analysis of corona discharge lacks consensus on the influence mechanism of oxygen--nitrogen components in air on its luminescent characteristics. This study aims to elucidate the underlying microphysical mechanisms through extensive experimental investigations. Based on our established high-time-resolution experiment platform for plasma imaging with quantum detection technology, we have conducted experiments on corona discharge under different nitrogen--oxygen ratios at a high time resolution. Our results show that in nitrogen, air, and oxygen environments, the luminescent features and development trends of corona discharge are generally consistent but exhibit differences in intensity and development speed depending on the gas environment. Upon analyzing both amplitude characteristics and temporal characteristics of the discharge luminescence, this study proposes an influence mechanism of oxygen components in air on corona discharge luminescence.