The aging of cross-linked polyethylene (XLPE) insulating materials directly affects the reliability and safety of power equipment. The traditional aging degree detection is mainly characterized by polarization-depolarization current

which lacks the characterization of the interface between macro and micro. So

phased array ultrasonic imaging technology is proposed to image XLPE samples with different thermal aging degrees. At the same time

a deep learning model between the phased array ultrasound imaging image and the aging degree is established

which realizes the efficient characterization of aging degree of XLPE. Through experiments

it is found that the differences of ultrasound image samples with different aging degrees are significant

and a certain degree of stratification phenomenon appears

which shows the feasibility to characterize the thermal aging process by ultrasonic array imaging

and the ultrasonic detection method has non-destructive properties

high sensitivity and good repeatability. The proposed method is not only of great significance to improve the operating safety of power system

but also provides a new perspective for the study of aging mechanism of XLPE insulating materials.