This paper presents a theoretical study relating to partial discharge (PD) source localization in high voltage power transformers using UHF technique. It also includes a brief review of typical PD sources, continuous UHF-based monitoring system, and some suitable UHF sensor types. Using MATLAB calculations and some basic assumptions, the differences in potential accuracy of PD source location are shown for different numbers of sensors, including the influence of sensor position. In addition, in the case of four UHF sensors, other effects were taken into accounts such as background noise sources, signal attenuation, multivariate wavelet denoising and the first peak detection method. By using the programs, at first, it is shown how to find the corresponding loci on which a PD source can be expected to lie by means of two and three UHF sensors in power transformers. Furthermore, using four sensors, it is presented on two different examples how simulated UHF signals, which travel from the partial discharge source to individual UHF sensors, can be used to find a partial discharge source position with sufficient accuracy. Multivariate wavelet transform was applied to the simulated signals in order to remove noise from the signals, and thus to reduce a localization error. The first peak detection method was applied to two various presumed shapes of UHF signal, in order to find time differences of arrival between related signals at the sensors. The corresponding results of partial discharge source location were analyzed and compared.

Application of UHF method for partial discharge source location in power transformers

The presence of metallic particles in oil insulation degrades the performance of power transformers. At present, a large amount of research has focused on PD and breakdown characteristics initiated by particles in stationary transformer oil. However, in actual use, transformer oil flows which affect the movement of particles, and hence PD and breakdown characteristics change. In order to clarify these, PD parameters and breakdown voltage were measured in moving transformer oil in the presence of metallic particles, and the effects of flow velocity, particle size and concentration were investigated. It is found that as the oil flow velocity increased, PDIV became higher, whereas the discharge magnitude and frequency decreased. On the contrary, the discharge magnitude and frequency increased with particle size and concentration. Besides, the breakdown voltage increased with the circulation of oil, but it became lower with increasing particle size and concentration. A simulation model including electric field and fluid field was employed to study the moving trajectory of particles in flowing transformer oil. When the oil flow velocity increased, the collision number between particles and electrodes decreased and it became larger with particle size and number. In terms of the simulation results, the variation of PD characteristics could be interpreted. The breakdown voltage, as affected by oil flow velocity, particle size and concentration were discussed.

Partial discharge and breakdown characteristics of moving transformer oil contaminated by metallic particles

Partial discharge (PD) detection and diagnosis based on the ultra-high frequency (UHF) signals is one of the most widely adopted methods to evaluate the internal insulation status of high voltage equipment. Benefit from the rapid development of computing hardware and data processing algorithms, the intelligent PD fault diagnosis method based on the UHF data has made considerable progress in the past two decades. This two-part paper aims to give a comprehensive review about the application of signal processing and machine learning technologies in UHF PD detection and diagnosis. These technologies are divided into three categories according to their respective purpose, which are the preprocessing technology, source localization technology and pattern recognition technology. As the first one of the two-part review, we focus on the preprocessing and localization approaches in this paper. Specifically, for the preprocessing topic, the methods for signal denoising, multi-source separation, and pulse segmentation are included. While for the localization topic, the time difference of arrival (TDOA) method, direction of arrival (DOA) method, received signal strength indicator (RSSI) method, and other latest methods are reviewed. For each topic, the basic ideas, recent research progresses, advantages and limitations are discussed in detail. Before the conclusion, we also make a discussion about the application effects of the above technologies and prospect some future directions accordingly. In the second paper, the pattern recognition problems based on the UHF PD data will be concentrated, especially the application of deep learning algorithms.

A Comprehensive Review of Signal Processing and Machine Learning Technologies for UHF PD Detection and Diagnosis (I): Preprocessing and Localization Approaches

Partial discharge (PD) detection has been proved as an effective tool for insulation condition monitoring of power equipment. The energy of PD pulses is valuable for studying the characteristics of PD activities. This paper proposes a method for estimating the energy of PD pulses in the presence of white noise and narrowband noise. First, a maximum likelihood (ML) estimator of the pulse energy is derived from the probability distribution of the energy spectral coefficients. To implement the ML method, the sampled data are divided into signal frames and noise frames. The noise frames are then utilized for extracting noise parameters using the 3F-C method. Eventually, these noise parameters are applied to the signal frames to find the ML estimate of the pulse energy. To verify the effectiveness of the proposed method, both simulated data and measured data have been processed using the proposed method and the conventional wavelet packet (WP) denoising method. Compared with the WP denoising method, the proposed method has a higher accuracy and is less susceptible to the lengths of the sampling time windows. The advantage of the method is more significant in unfavorable conditions where the signal-to-noise ratio is low and the accurate lengths of the PD pulses are difficult to determine.

Energy Estimation of Partial Discharge Pulse Signals Based on Noise Parameters

During the operation of power transformers, the insulating oil is actually in a moving state due to the effects of various cooling systems. Therefore, the investigation on insulation characters of moving transformer oil is necessary. In this paper, the influences of temperature on partial discharge (PD) and breakdown characteristics of moving transformer oil contaminated by metallic particles were experimentally studied at first. It was found that the PD magnitude and frequency first decreased and then became higher when the temperature increased from 40 °C to 80 °C, and a minimum point appeared between 70 °C and 80 °C. On the contrary, the dependence of breakdown voltage on temperature showed an opposite tendency. In order to explain the experimental results, a simulation model involving multiple forces was constructed to obtain the movement trajectory of particles. The increase of oil temperature leaded to the decrease of dynamic viscosity of transformer oil, so the forces acting on particles and their movement characters changed. As the temperature increased, the collision frequency between particles and electrodes first decreased and then became higher, and there was a minimum point between 70 °C and 80 °C. This changing tendency was in accordance with the dependence of PD parameters on the oil temperature. In terms of the above results, the effects of temperature on PD frequency and breakdown voltage were discussed.

Variation of Discharge Characteristics With Temperature in Moving Transformer Oil Contaminated by Metallic Particles

Metal particle contamination can cause partial discharge (PD) and seriously degrade the withstand voltage of transformer oil. The moving metal particles, which display PD characteristics that differ from those of free particles, perform complex movements within the transformer oil duct under the force of a submerged pump. This study analyzed the Φ-U-N diagram and measured the inception voltage, number, and amplitude of PD under different supply voltages. The influence of flow velocity on the characteristic parameters of PD was examined. The forces of charged metal particles were analyzed, and the trajectories were simulated in an oil duct under different flow velocities. As the supply voltage was increased, the number and amplitude of PD increased, whereas the shape of the Φ-U-N diagram remained the same. As the flow velocity was increased, the inception voltage of PD increased, but the number and amplitude of PD decreased. The ΦU-N diagram demonstrated slight changes under different flow velocities. Overall, an increase in flow velocity reduces PD possibly because of particle oscillation within the system.

Impact of velocity on partial discharge characteristics of moving metal particles in transformer oil using UHF technique

In view of the phenomenon that the internal medium of power electronic devices applying immersion phase change cooling technology has been boiling during operation, it is necessary to carry out research on whether long-term boiling of the medium will lead to the pyrolysis of the medium and the pyrolysis process and products when the medium is pyrolyzed. In this paper, 25 FC-72 simulated molecular models are set up in the ReaxFF force field by molecular dynamics simulation method to study the dynamic pyrolysis process of FC-72 at different temperatures, and the following conclusions are drawn by analyzing the pyrolysis products: the pyrolysis process mainly generates some small molecule gases such as CF4 and F2, and the reaction process is significantly accelerated with the increase of temperature, the higher the temperature, the more C6F14 participates in pyrolysis, and the faster the reaction system reaches stability.

Jinsheng Su

Papers

Impact of velocity on partial discharge characteristics of moving metal particles in transformer oil using UHF technique

Phase Change Cooled Encapsulation Material FC-72 Pyrolysis Process Simulation