While frequency response analysis (FRA) is a well matured technique widely used by current industry practice to detect the mechanical integrity of power transformers, interpretation of FRA signatures is still challenging, regardless of the research efforts in this area. This paper presents a method for reliable quantitative and qualitative analysis to the transformer FRA signatures based on a decision tree classification model and a fully connected neural network. Several levels of different six fault types are obtained using a lumped parameter-based transformer model. Results show that the proposed model performs well in the training and the validation stages, and is of good generalization ability.

Air-core coil current transformer is a key piece of equipment in the digital substation development. However, it is more vulnerable to various faults when compared with the traditional electromagnetic current transformer. Aiming at understanding the effect of various parameters on the performance of the air-core coil current transformer, this paper investigates the influence of these factors using the maximum information coefficient. The interference mechanism of influencing factors on the transformer error is also analyzed. Finally, the Stacking model fusion algorithm is used to predict transformer errors. The developed base model consists of deep learning, integrated learning and traditional learning algorithms. Compared with gated recurrent units and extreme gradient boosting algorithms, the prediction model based on stacking model fusion algorithm proposed in this paper features higher accuracy and reliability which helps improve the performance and safety of future digital substations.

Winding fault is one of the most common types of transformer faults. The frequency response method is a common diagnosis method for winding fault detection. In order to improve the feature extraction ability of the frequency response curve before and after the winding fault, this paper proposes a winding fault feature extraction method based on the moving window algorithm to improve the Euclidean distance and correlation coefficient and uses a support vector machine to diagnose winding fault. “Moving window meter algorithm” refers to the fixed moving window width and window moving interval, scanning the entire frequency response curve from the initial point to the end point of the frequency response curve, using the correlation coefficient (CC) and Euclidean distance (ED) to calculate the mathematical index of each window. The mathematical index of each window is used as the characteristic quantity of fault type classification. Finally, the grid search algorithm is used to optimize the support vector machine to classify and identify the type of winding fault. At the same time, the standard support vector machine s(SVM) and back propagation neural network algorithm (BPNN) are compared with the support vector machine optimized by the grid search method to diagnose the fault type. The research shows that the improved correlation coefficient and Euclidean distance using the moving window algorithm are more sensitive to winding faults than the traditional calculation methods. The combination of the two calculation methods makes up for the shortcomings of their respective methods. The fault features obtained meet the requirements of the support vector machine for fault diagnosis, and the grid search method-optimized support vector machine classification algorithm has a good classification and recognition effect on the identification of fault types. The effectiveness and superiority of this method are further illustrated.

High-voltage switchgears can generate high-frequency and high-amplitude electromagnetic interference during their operation, which may lead to abnormal outputs of the measurement devices, thus giving rise to measurement errors or protective relay malfunctions and impacts on the stable running of the power grid. Considering the air breakdown delay, a dynamic reignition model consisting of the hyperbolic model, the static model and the improved Mayr model is proposed to analyze the transient process during the on/off operation of open-type disconnectors. Meanwhile, the influence of arc resistance, line length and breakdown delay on the model were analyzed, and an experimental circuit was built to verify the simulation results through a comparison. It was found that the proposed model could reflect the high-frequency voltage and current during the on/off operation more accurately and thus provide an effective method for the precise modeling of the switching operation of the disconnector.