Transformer fault diagnosis method based on deep learning

Abstract

The invention discloses a transformer fault diagnosis method based on deep learning, and belongs to the field of transformer fault diagnosis. The method comprises the following steps: firstly, carrying out duplicate removal and abnormal value processing on concentration data of fault characteristic gases H2, CH4, C2H2, C2H4 and C2H6 collected by an analysis method of dissolved gas in oil, fillingmissing values by using a random forest method, and then carrying out normalization processing on the data to form a training set sample and a test set sample; establishing a three-layer stack type sparse noise reduction auto-encoder model, and rewriting a cross entropy loss function in a traditional classification model into a Focal loss function; according to the method, hyper-parameters are determined through class sample weights, white Gaussian noise is added into input, an auto-encoder is made to fully extract effective features, and therefore an effective feature extraction model is obtained, and a Softmax classifier is used for outputting a diagnosis result of the model. Compared with the existing methods such as a three-ratio method, an SVM and a BP neural network, the transformerfault diagnosis method provided by the invention has good diagnosis performance, and the accuracy of transformer fault diagnosis is effectively improved.

Description

technical field [0001] The invention belongs to the field of transformer fault diagnosis and relates to a transformer fault diagnosis method based on deep learning. Background technique [0002] Transformer is the core equipment of power system operation, and accurate diagnosis of latent faults inside transformer is of great significance to the safe operation of power grid. Dissolved gas analysis (DGA) in oil is an effective method for diagnosing and detecting latent faults inside transformers. The fault diagnosis methods of power transformers based on DGA are mainly divided into traditional fault diagnosis methods and intelligent diagnosis methods. [0003] The traditional methods mainly include the three-ratio method and the improved three-ratio method. The basic principle of the three-ratio method is that when a transformer fails, the corresponding three-ratio value is calculated from the characteristic gas content extracted from the transformer oil and given a correspon...

AI Technical Summary

Problems solved by technology

However, the above method has gradually revealed shortcomings such as incomplete coding and too absolute judgment standards in the practice process.

[0004] Intelligent methods mainly include expert system, support vector machine, fuzzy theory method, artificial neural network method, etc. These existing intelligent methods have improved the accuracy of transformer fault diagnosis to varying degrees, but there are still some problems and certain limitations.

For example, the expert system requires a large amount of correct expert experience, and it is difficult to apply in practice; the support vector machine method is essentially a binary classification algorithm, and transformer fault diagnosis is a multi-classification problem. In the face of multi-classification problems, the parameter setting is cumbersome, and the process of constructing a classifier is cumbersome; The fuzzy theory method needs to artificially set the initial clustering center, and the diagnostic effect is greatly limited by the initial clustering center; the neural network method has the defects of slow convergence speed and easy to fall into the local optimal solution

Images