Research on the Detection Method of Transformer Oil Performance Parameters Based on Multi-frequency Ultrasonic Technology and IMFO-BP

摘要

Transformer oil is an important part of transformer insulation, and some important performance parameters of transformer oil need to be quickly and accurately tested. A transformer oil performance detection model based on multifrequency ultrasonic (MFU) technology and improved Moth-Flame optimization algorithm (IMFO) optimized Back Propagation (BP) Neural Network is proposed. Let the ultrasonic waves of multiple frequencies pass through the transformer oil, and measure the transmission speed and attenuation coefficient at different frequencies. Introduce adaptive weight and Levy flight to improve the Moth-Flame optimization algorithm to optimize the BP neural network, and use the IMFO-BP algorithm with strong optimization ability and high precision to train the samples to obtain the transformer oil performance parameter detection model. Verify its feasibility through experimental testing.