A power transformer fault diagnosis method based on acoustic characteristics and a neural network

Abstract

The invention discloses a power transformer fault diagnosis method based on acoustic characteristics and a neural network, and the method comprises the following steps: employing a sound collection device to collect and obtain sound signals when a power transformer is in each state, and recording the corresponding relation between the collected sound signals and each state of the power transformer; preprocessing the acquired sound signals; Establishing and training a GRU neural network model; and collecting a sound signal of the power transformer to be diagnosed, preprocessing the sound signal, inputting the preprocessed sound signal into the trained GRU neural network model, and completing fault diagnosis of the power transformer to be diagnosed according to an output result of the GRU neural network model. According to the method, the frequency domain characteristics of the power transformer are extracted from the sound signals generated when the power transformer runs, the frequencydomain characteristics of the power transformer are used for training the threshold circulation unit neural network, operation is relatively simple, cost is low, and online monitoring is easy to achieve.

Description

technical field [0001] The invention belongs to the technical field of power transformer fault diagnosis, and in particular relates to a power transformer fault diagnosis method based on acoustic features and a neural network. Background technique [0002] As one of the important equipment in the power system, the power transformer undertakes key tasks such as voltage conversion, power distribution and transmission inside the power system. During the operation of power transformers, faults such as discharge, overheating, insulation degradation, loose winding and iron core, and solid pollution of insulating oil may occur. In-depth study of power transformer fault diagnosis method is of great significance to the stable operation of power system. [0003] With the continuous development and improvement of machine learning theory, the nonlinear mapping ability, self-learning ability and fault tolerance ability of neural network are continuously enhanced, and the application of ...

AI Technical Summary

Problems solved by technology

In the practical application of this solution, contact or non-contact measurement of the dissolved gas content in the power

Images