Dissolved gas analysis (DGA) is widely used to detect the incipient fault of power transformers However, the accuracy is greatly limited by selection of DGA features and performance of fault diagnostic model This paper proposed a fault diagnostic method integrating feature selection and diagnostic model optimization Firstly, this paper set up three feature sets with eight basic DGA gases, 28 DGA gas ratios and 36 hybrid DGA features, respectively Then, to eliminate the interference of weak-relevant and irrelevant features, the genetic-algorithm-SVM-feature-screen (GA-SVM-FS) model was built to screen out the optimal hybrid DGA features subset (OHFS) from three feature sets Next, using the OHFS as the input, the support vector machine (SVM) multi-classifier optimized by ISGOSVM (SVM classifier optimized by improved social group optimization) was built to diagnose fault types of transformers Finally, the performance of OHFS and ISGOSVM diagnostic model was tested and compared with traditional DGA features and diagnostic models, respectively The results show that the OHFS screened out is comprised of 14 features, including 12 gas ratios and two gases The accuracy of OHFS is 3–30% higher than traditional DGA features, and the accuracy of ISGOSVM can increase by 3% to 14% compared with the SGOSVM (SVM classifier optimized by social group optimization), GASVM (SVM classifier optimized by genetic algorithm optimization), PSOSVM (SVM classifier optimized by particle swarm optimization), and SVM diagnostic models The proposed approach integrating the OHFS with ISGOSVM achieves the highest accuracy of fault diagnose (9286%)

A Transformer Fault Diagnosis Model Using an Optimal Hybrid Dissolved Gas Analysis Features Subset with Improved Social Group Optimization-Support Vector Machine Classifier

Early detection and localization of winding deformation is a major task in ensuring proper condition assessment and maintenance of transformers. This work deals with the formulation of a generalized approach by finite element method (FEM) based analysis on magnetic flux distribution for finding the presence of inter-disc fault and calculation of severity and location of faults in the winding. Artificial short circuits of various severity are created in the simulation model of transformer in ANSYS software which solves the magnetic field intensity and flux density around the winding geometry by finite element method (FEM). The detailed investigation on the simulation results identifies parameters based on the percentage changes in the field intensity and flux density at symmetrical discs in case of short circuits in the winding. The parameters based on the changes in peak values of magnetic field intensity and flux density at symmetrical discs over a cycle identified to be suitable for detection and localization of faults with respect to centre of winding. Moreover, a parameter based on the changes in instantaneous values of flux density at symmetrical discs proves to be suitable for discriminating the faults at upper and lower halves of the winding.

Development of a Magnetic Field Measurement Based Scheme for Detection and Localization of Inter-Disc Faults in Transformer Windings by Finite Element Method

Frequency response analysis (FRA) is one of the most reliable and useful methods for mechanical fault detection in transformers. The winding axial displacement is one of the mechanical fault types that can occur in transformers. Axial displacements causes the detailed model parameters changing, in response to which FRA characteristics will also change. Numerous researchers have considered winding axial displacements as a symmetrical movement of disks. However, one of the most prevalent types of winding axial displacements is asymmetrical axial displacement which has not been considered by any prior studies. In this paper, physical asymmetrical and symmetrical axial displacements of the first disk of the LV winding were simulated and FRA characteristics were obtained for these cases. Finally, experimental tests were performed on an experimental transformer to validate the simulation results.

Effect of the asymmetrical axial displacement of transformer windings on fra characteristics

https://www.ijew.io/pdfdownload/high-frequency-modeling-of-transformer-using-black-box-frequency-response-analysis

High Frequency Modeling of Transformer Using Black Box Frequency Response Analysis

Reliability of Active Front End (AFE) converter can be improved if converter faults can be identified before startup. Startup diagnostics of the LCL filter can be an useful tool to accomplish this. An algorithm based on Fast Fourier Transform (FFT) of the filter step response is shown to be able to detect variation of filter components. This is extended to consider conditions of short circuit and open circuit of filter components. This method can identify failed components in a LCL filter before operating the converter, which otherwise may lead to undesirable operation. Analytical expressions are derived for the frequency spectrum of the LCL filter during step excitation using continuous Fourier Transform and Discrete Fourier Transform (DFT). A finite state machine Finite State Machine (FSM) based algorithm is use to sequence the startup diagnostics before commencing normal operation of the power converter. It is shown that the additional computational resource required to perform the diagnostic algorithm is small when compared with the overall inverter control program. The diagnostic functions can be readily implemented in advanced digital controllers that are available today. The spectral analysis is supported by simulations and experimental results which validate the proposed method.

LCL filter startup diagnostics for an Active Front End converter

Power transformer is the most expensive single element of HV transmission system. Partial Discharge, mechanical displacement of the winding and inter-turn fault is commonly encountered problem in power transformer. This minor fault may lead to irretrievable damage. Transfer Function Method and Frequency Response Analysis Method are used in this paper, in order to locate the winding fault. The end result obtained for the different fault condition is compared with the reference set and the conclusions are drawn. Simulation is done with Orcad-Pspice and its results are verified with experimental setup.

Paulraj. T

Papers

Modeling and location of faults in power transformer using Transfer Function and Frequency Response Analysis