Simulation study on nondestructive testing of winding hot spot temperature based on ultrasonic emission

The rise in hot spot temperature is a critical indicator of transformer winding defects, making its monitoring essential. Traditional temperature measurement methods, however, have certain constraints in obtaining hot spot information due to the complex structure of transformers. This article thus proposes a nondestructive testing for measuring the transformer winding hot spot temperature based on ultrasonic thermometry, without altering the transformer’s internal structure. It also explores the potential application of ultrasonic temperature measurement technology inside transformers. The propagation pattern of ultrasound within a transformer is simulated, and the effect of hot spot faults at varying positions and temperatures on the spatiotemporal distribution of ultrasonic propagation is examined. The range of effective acoustic signals has been determined through comparison. An artificial intelligence-based hot spot temperature inversion method is proposed, leading to the establishment of an effective diagnostic technique for hot spot temperature within the hot spot region, accomplishing a preliminary inversion of hot spot faults. This research offers a novel approach to measuring the internal temperature of transformers.