641 / 2022-03-30 22:52:37

Detection of dielectric loss factor of transformer oil based on multi frequency ultrasonic detection technology and WOA-Elman

dielectric loss factor,transformer oil,multi frequency ultrasonic,WOA-Elman,detection model

The dielectric loss factor is a very important index for evaluating the insulation performance of transformer oil, which is very sensitive to the deterioration of oil and the degree of pollution. Ultrasound detection technology obtains information by identifying changes in the characteristic parameters of ultrasound in a medium. As a nondestructive testing technology, ultrasonic can identify polar impurities through attenuation characteristics, and then determine dielectric loss of transformer oil. In this paper, ultrasonic reflection detection method and ultrasonic transmission detection method are used to detect transformer oil, and 242 dimensional multi frequency ultrasonic data set is obtained. WOA-Elman detection model is established in Matlab environment, with ultrasonic data set as input and dielectric loss factor as output for training and prediction. Among 175 groups of transformer oil samples, 160 groups are used for model training and 15 groups are used for prediction. The results show that the prediction accuracy of transformer oil dielectric loss model based on WOA Elman is 93.2%, which proves that multi frequency ultrasonic detection combined with artificial intelligence model can effectively detect transformer oil dielectric loss.