Anti-Interference Deep Visual Identification Method for Fault Localization of Transformer Using a Winding Model

Abstract

The idea of Ubiquitous Power Internet of Things (UPIoTs) accelerates the development of intelligent monitoring and diagnostic technologies. In this paper, a diagnostic method suitable for power equipment in an interference environment was proposed based on the deep Convolutional Neural Network (CNN): MobileNet-V2 and Digital Image Processing (DIP) methods to conduct fault identification process: including fault type classification and fault localization. A data visualization theory was put forward in this paper, which was applied in frequency response (FR) curves of transformer to obtain dataset. After the image augmentation process, the dataset was input into the deep CNN: MobileNet-V2 for training procedures. Then a spatial-probabilistic mapping relationship was established based on traditional Frequency Response Analysis (FRA) fault diagnostic method. Each image in the dataset was compared with the fingerprint values to get traditional diagnosing results. Next, the anti-interference abilities of the proposed CNN-DIP method were compared with that of the traditional one while the magnitude of the interference gradually increased. Finally, the fault tolerance of the proposed method was verified by further analyzing the deviations between the wrong diagnosing results with the corresponding actual labels. Experimental results showed that the proposed deep visual identification (CNN-DIP) method has a higher diagnosing accuracy, a stronger anti-interference ability and a better fault tolerance.

Classification of Catheters and Tubes on Chest Radiographs Using Light-Weight Deep Convolutional Neural Networks

“…With expanding numbers of AI models being implemented in the production environment, the system's burden is rapidly growing [23]. However, a model with high diagnostic accuracy and low calculation processing resource consumption is the cornerstone of the integrated medical AI system [24]. In the testing dataset, we employed the CPU to simulate the limited computing resources in the production environment.…”

Classifying transformer winding deformation type by combination of FRA polar plot texture feature and multiple SVM classifiers

“…The above research confirms that the polar plot integrates the amplitude and phase characteristics of frequency response information, which improves the sensitivity of fault diagnosis compared to the traditional FRA method. However, in the above research, apart from RD faults, there was no discussion on other common winding deformation faults, such as cake to cake faults and cake to cake short circuit faults; More importantly, the article did not provide specific diagnostic strategies and methods [17][18][19] . The texture features of frequency response polar maps of several common transformer faults have been studied [20] , which is of great significance for fault diagnosis.…”