The article presents a novel on-line partial discharge (PD) monitoring system for power transformers, whose functioning is based on the simultaneous use of three unconventional methods of PD detection: high-frequency (HF), ultra-high frequency (UHF), and acoustic emission (AE). It is the first monitoring system equipped in an active dielectric window (ADW), which is a combined ultrasonic and electromagnetic PD sensor. The article discusses in detail the process of designing and building individual modules of hardware and software layers of the system, wherein the most attention was paid to the PD sensors, i.e., meandered planar inverted-F antenna (MPIFA), high-frequency current transformer (HFCT), and active dielectric window with ultrasonic transducer, which were optimized for detection of PDs occurring in oil-paper insulation. The prototype of the hybrid monitoring system was first checked on a 330 MVA large power transformer during the induced voltage test with partial discharge measurement (IVPD). Next, it was installed on a 31.5 MVA substation power transformer and integrated according to the standard IEC 61850 with SCADA (Supervisory Control and Data Acquisition) system registering voltage, active power, and oil temperature of the monitored unit. The obtained results showed high sensitivity of the manufactured PD sensors as well as the advantages of the simultaneous use of three techniques of PD detection and the possibility of discharge parameter correlation with other power transformer parameters.

This article describes the design of UHF Hilbert curve fractal antenna (HCFA) specially adapted for the partial discharge monitoring system. The authors present the mathematical apparatus for calculating resonant frequencies of Hilbert fractal antenna and results of a computer simulation of the developed prototype. In the design process, the antenna’s working environment (mineral oil) and the mechanical construction of the transformer inspection window were taken into consideration as well. The article also shows the results of laboratory tests carried out in the transformer tank model with different type of partial discharge sources. Both simulations and partial discharge measurements showed that the HCFA, due to such properties as: multi-resonance, small size, low fabrication cost and high sensitivity, is an interesting alternative to other UHF probes installed in transformer inspection window

The article presents research on the influence of the type of UHF antenna and the type of machine learning algorithm on the effectiveness of classification of partial discharges (PD) occurring in the insulation system of a power transformer. For this purpose, four antennas specially adapted to be installed in the transformer tank (UHF disk sensor, UHF drain valve sensor, planar inverted F-type antenna, Hilbert curve fractal antenna) and a reference log-periodic antenna were used in laboratory tests. During the research, the main types of PD, typical for oil-paper insulation, were generated, i.e., PD in oil, PD in oil wedge, PD in gas bubbles, surface discharges, and creeping sparks. For the registered UHF PD pulses, nine features in the frequency domain and four features in the wavelet domain were extracted. Then, the PD classification process was carried out with the use of selected methods of supervised machine learning. The study investigated the influence of the number and type of feature on the obtained classification results gained with the following machine-learning methods: decision tree, support vector machine, Bayes method, k-nearest neighbor, linear discriminant, and ensemble machine. As a result of the works carried out, it was found that the highest accuracies are gathered for the feature representing peak frequency using a decision tree, reaching values, depending on the type of antenna, from 89.7% to 100%, with an average of 96.8%. In addition, it was found that the MRMR method reduces the number of features from 13 to 1 while maintaining very high effectiveness. The broadband log-periodic antenna ensured the highest average efficiency (100%) in the PD classification. In the case of the tested antennas adapted to work in an energy transformer tank, the highest defect-recognition efficiency is provided by the UHF disk sensor (99.3%), and the lowest (89.7%) is by the UHF drain valve sensor.

Wydawnictwo SIGMA-NOT wydaje czasopisma fachowe informuj?ce swoich czytelnik?w o najnowszych osi?gni?ciach naukowych i nowoczesnych rozwi?zaniach technicznych w Polsce i na ?wiecie, popularyzuje problemy techniczne oraz poszerza wiedz? i kultur? techniczn?.