Abstract:Since the traditional transformer fault diagnosis method based on dissolved gas analysis (DGA) is challenging to meet today's engineering needs, this paper proposes a multi-model fusion transformer fault diagnosis method based on TimesNet and Informer. First, the original TimesNet structure is improved by adding the MCA module to the Inception structure of the original TimesBlock to reduce the model complexity and computational burden; second, the MUSE attention mechanism is introduced into the original TimesNet to act as a bridge, so that associations can be carried out effectively among the local features, thus enhancing the modeling capability of the model; finally, when constructing the feature module, the TimesNet and Informer multilevel parallel feature extraction modules are introduced, making full use of the local features of the convolution and the global correlation of the attention mechanism module for feature summarization, so that the model learns more time-series information. To verify the effectiveness of the proposed method, the model is trained and tested on the public DGA dataset, and the model is compared and experimented with classical models such as Informer and Transformer. The experimental results show that the model has a strong learning ability for transformer fault data and has an advantage in accuracy compared with other models, which can provide a reference for transformer fault diagnosis.

What problem does this paper attempt to address?

The paper attempts to address the challenges in transformer fault diagnosis. Traditional transformer fault diagnosis methods based on Dissolved Gas Analysis (DGA) are difficult to meet modern engineering needs because the internal faults of transformers are very complex, and the traditional statistical analysis coding combinations are too absolute and rough, leading to low diagnostic accuracy and even possible misjudgment or omission. In addition, although existing time series analysis methods can extract features from different perspectives, their way of creating time features is relatively single, making it difficult to meet the needs of various time series tasks. To solve these problems, the paper proposes a multi-model fusion transformer fault diagnosis method based on TimesNet and Informer. This method improves the learning ability and diagnostic accuracy of the model for transformer fault data by improving the TimesNet structure, introducing the MUSE attention mechanism, and constructing a multi-level parallel feature extraction module. Specifically: 1. **Improving the TimesNet structure**: By adding the MCA module to the Inception structure of TimesNet, the complexity and computational burden of the model are reduced. 2. **Introducing the MUSE attention mechanism**: As a bridge, it effectively associates local features and enhances the modeling ability of the model. 3. **Constructing a multi-level parallel feature extraction module**: Combining the multi-level parallel feature extraction modules of TimesNet and Informer, it fully utilizes the local features of convolution and the global correlation of the attention mechanism module, enabling the model to learn more temporal information. Through these improvements, the paper aims to improve the accuracy and efficiency of transformer fault diagnosis, providing a new reference method for transformer fault diagnosis.

Transformer Fault Diagnosis Method Based on TimesNet and Informer

Research On Neural Network Fault Diagnosis Method Based On Dga For Large Transformer

A Methodology for Transformer Fault Diagnosis Based on the Feature Extraction from DGA Data

A Feature Selection and Ensemble Learning Based Methodology for Transformer Fault Diagnosis

A Methodology to Diagnose Transformer Faults Based on Principal Components Analysis and Artificial Neural Network

Generalized Transformer in Fault Diagnosis of Tennessee Eastman Process

Application of a Combinatorial Neural Network Model Based on Cluster Analysis in Transformer Fault Diagnosis

Multi-filter semi-supervised transformer model for fault diagnosis

A Novel Neural Network Approach to Transformer Fault Diagnosis Based on Momentum-Embedded BP Neural Network Optimized by Genetic Algorithm and Fuzzy c-Means

Convolutional Bi-directional Long Short Term Memory Network based Dynamic Fault Diagnosis for Transformer DGA

Fault Prediction Using a Grey-Markov Model from the Dissolved Gases Contents in Transformer Oils

Power Transformer Fault Diagnosis using Dynamic Multiscale Graph Modeling and M2SGCN Network Based on Statistical Fusion

Gaussian-Linearized Transformer with Tranquilized Time-Series Decomposition Methods for Fault Diagnosis and Forecasting of Methane Gas Sensor Arrays

A Methodology to Diagnose Transformer Faults Based on the Multi-Objective Optimization and the Case-Based Reasoning

Convolutional neural networks applied to dissolved gas analysis for power transformers condition monitoring

Fault diagnosis method for oil-immersed transformers integrated digital twin model

An Integrated Method of ANFIS and Dempster-Shafer Theory for Fault Diagnosis of Power Transformer

Transformer fault diagnosis model based on FI-CNN method

Online Fault Diagnosis of Power Transformer with Temporal and Spatial Features

Transformer fault diagnosis method based on SMOTE and NGO-GBDT

A New Method for Transformer Fault Prediction Based on Multifeature Enhancement and Refined Long Short-Term Memory