INFLUENCE OF THERMAL STRESS RATE ON CHANGES IN ELECTRICAL INSULATING OILS

This work solves the comparison of two different thermal stress times for selected liquid insulation materials based on transformer oils by impedance spectroscopy. Research is focused on actual and progressive liquid insulation materials. The scientific objectives of the experiment are focused on the determination of hypotheses, defined by predictable electrophysical parameters in experimental conditions, which are described in the analysis of the measured results.

Determination of Transformer Oil Quality Parameters by its Color Coordinates

The service life of transformers and, as a consequence, of the electric power systems depends largely on the chemical composition and electrical insulating properties of the oil. In this regard, much attention is paid to the control of physical and chemical indicators and operational properties of transformer oils. The aim of this work is to determine the important diagnostic parameters such as the acid number and the tangent of the dielectric loss angle of the transformer oil based on the measured coordinates of the color of the transformer oil. This goal is achieved by the study of the coordinates of the transformer oils chromaticity. For this purpose, a white light source is used, which is a LED with a pumping line of 450 nm and a maximum radiation in the region of 550-560 nm, with radiation in the range from 400 to 800 nm. Sensors 1, 2, 3 allow determining the chromaticity coordinates of the measured source R, G, B (red, green, blue), due to several receivers having spectral characteristics of sensitivity in the corresponding region of the visible spectrum. The most significant result of the work was the establishment of correlations between the TM chromaticity coordinatesб their acid number and the tangent of the dielectric loss angle of TM. The significance of the results obtained was that the acid number and the tangent of the dielectric loss angle of the transformer oil could be determined by the chromaticity coordinates, and hence, by the color of the oil.

Investigation on the Dielectric Properties of Palm Oil with Silicon Carbide Doping for Transformer Application

The dielectric strength of insulating liquids of transformer acts an important parameter in the operation of transformer. Thus, great interest and many studies have been extensively done to improve the dielectric strength. One of study is the introduction of nanoparticle in the transformer oils. Study of the nanoparticles for the last few years had been found that, it can be dispersed in the transformers oils to be nanofluids and directly enhance the transformer performance. In this study, an investigation has been carried out to focus on the effect of silicon carbide (SiC) nanoparticle to AC (alternating current) breakdown voltage of the Refined, Bleached, and Deodorized Palm Oil (RBDPO). AC breakdown test have been conduct according to the standard of the IEC 60156. Besides that, a number of parameters will be evaluated such as dielectric dissipation factor (tan δ), relative permittivity (ε), and resistivity (ρ). Based on the results of the experiment, the electrode gap at 2.5 mm having the highest AC breakdown voltage compared to the other electrode gap which are 1.0 mm, 1.5 mm and 2.0 mm. Furthermore, doping with different concentrations of the silicon carbide (SiC) in Refined, Bleached, and Deodorized Palm Oil (RBDPO) found decreasing of AC breakdown voltage from 52.09 kV (without SiC) to 45.3 kV for 0.001 g/L, 43.2 kV for 0.003 g/L and 40.1 kV for 0.005 g/L respectively.

Transformer Oil Quality Assessment Using Random Forest with Feature Engineering

Machine learning is widely used as a panacea in many engineering applications including the condition assessment of power transformers. Most statistics attribute the main cause of transformer failure to insulation degradation. Thus, a new, simple, and effective machine-learning approach was proposed to monitor the condition of transformer oils based on some aging indicators. The proposed approach was used to compare the performance of two machine-learning classifiers: J48 decision tree and random forest. The service-aged transformer oils were classified into four groups: the oils that can be maintained in service, the oils that should be reconditioned or filtered, the oils that should be reclaimed, and the oils that must be discarded. From the two algorithms, random forest exhibited a better performance and high accuracy with only a small amount of data. Good performance was achieved through not only the application of the proposed algorithm but also the approach of data preprocessing. Before feeding the classification model, the available data were transformed using the simple k-means method. Subsequently, the obtained data were filtered through correlation-based feature selection (CFsSubset). The resulting features were again retransformed by conducting the principal component analysis and were passed through the CFsSubset filter. The transformation and filtration of the data improved the classification performance of the adopted algorithms, especially random forest. Another advantage of the proposed method is the decrease in the number of the datasets required for the condition assessment of transformer oils, which is valuable for transformer condition monitoring.