Search

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. [ABSTRACT FROM AUTHOR]

The research reported in this paper involves the development and refinement of methods applicable to the measurement and analysis of infrasound signals generated by the operation of wind turbines. In particular, the presentation focuses on the use of a new system that is applied for simultaneous recording of acoustic signals in the low-frequency range emitted by wind farms in three independent and identical measurement setups. A comparative analysis of the proposed new system was made with the Brüel & Kjaer measurement, a commonly used methodology, which meets the requirements of the IEC 61400-11 standard. The paper focuses on the results of frequency and time-frequency analysis of infrasound signals recorded throughout the operation of a wind turbine with a rated capacity of 2 MW. The use of a correlated system with three simultaneous measurement systems can be a new and alternative measurement method that will eliminate the drawbacks of previous approaches. [ABSTRACT FROM AUTHOR]

The development of wind energy and the increasing number of installed wind turbines make it necessary to assess them in terms of the nuisance of the emitted infrasound noise generated by such devices. The article presents the results of measurements and analyses of infrasound emitted during the operation of wind turbines installed in various locations in Poland. Comparative analysis of noise levels in the infrasound and audible range has shown that acoustic energy is mainly in the low and infrasound frequency range, and the measured levels depend significantly on the weighting curves used. On the basis of the results, it was confirmed that the sound pressure level of infrasound signals emitted by the operation of high-power wind turbines, regardless of wind velocity, weather conditions, design solutions of turbines, operating time, rated capacity, does not exceed the criteria specified in the applicable legislation dealing with the assessment of infrasound noise on the working environment. [ABSTRACT FROM AUTHOR]