Power transformer online detection system based on improved convolutional neural network, and method

Abstract

The invention provides a power transformer online detection system based on an improved convolutional neural network, and a method. The online detection system specifically comprises a sensor set, a signal amplification device, a signal collecting and storing device and a fault positioning device; and the method for the power transformer online detection system based on the improved convolutionalneural network comprises the steps of offline training part and online detection part. According to structural characteristics and the vibration principle of a transformer, distribution points of a vibration sensor are determined, and transformer vibration signals are obtained in real time; original signals are subjected to data processing to be converted into a two-dimensional grey-scale map to serve as convolutional neural network input, wherein the two-dimensional grey-scale map is easily identified through the convolutional neural network; and according to the characteristic that noise interference of real-time vibration data of the transformer is large, the structure of a traditional convolutional neural network is improved, the size of a convolution kernel is improved, an improved convolutional neural network framework applied to power transformer fault positioning is built, and reasonability and superiority of the method are verified.

Description

technical field [0001] The invention belongs to the technical field of power transformer fault diagnosis, and in particular relates to an online power transformer detection system and method based on an improved convolutional neural network. Background technique [0002] As the key equipment for power system safety, power transformer is also one of the most expensive and complicated equipment. Since the power transformer is in the key position of the power grid and the operating environment is complicated, once a failure occurs, it will largely cause grid failures, large-scale power outages, etc., and even cause serious accidents such as explosions and fires. The direct or indirect economic loss can be as high as hundreds of millions of RMB. Therefore, whether the power transformer can operate safely, reliably and stably is closely related to the normal production and life of the people. With the increase of power grid capacity and the concept of "smart grid", smart substat...

AI Technical Summary

Problems solved by technology

Since the power transformer is in the key position of the power grid and the operating environment is complicated, once a failure occurs, it will largely cause grid failures, large-scale power outages, etc., and even cause serious accidents such as explosions and fires. The direct or indirect economic loss can be as high as hundreds of millions of RMB. Therefore, whether the power transformer can operate safely, reliably and stably is closely related to the normal production and life of the people.

Images