فهرست نشریات علمی دسترسی آزاد ایران

The condition assessment of silicon rubber insulators has always been a crucial requirement in power systems. Utilizing leakage current as a fast and online method has played a significant role in this regard. However, variations in operating conditions can lead to substantial errors in decisions regarding insulator condition. Under the influence of environmental conditions on the leakage current, it is difficult to define criteria capable of distinguishing different operating conditions. In this regard, achieving a model with minimal complexity for decision-making is considered as a suitable solution. In this study, an index based on leakage current is presented to categorize the performance of insulators into three states: normal, caution, and critical. For this purpose, experimental data were practically collected under light, medium, and heavy pollution conditions in varying humidity levels up to 90%. All tests were performed under different environmental conditions on healthy and aged insulators to investigate the effect of surface degradation on leakage current and their harmonic components. To analyze the leakage current, wavelet transform was employed, and the standard deviation of wavelet detail coefficients was used. Also, distinguishing a healthy insulator from one that has been aged by UV radiation based on detail coefficients is introduced in this paper. From the wavelet transform results, it appears that a D6 standard deviation above 0.01 reliably separates healthy insulators from aged ones. Furthermore, by using the standard deviation of D7 and defining threshold limits for healthy and aged insulators, it became possible to differentiate their performance. In order to evaluate the accuracy and sensitivity of the proposed indicators, the confusion matrix was used, which is able to distinguish different operating conditions with an accuracy of 89.58%.