Performance Detection of Transformer Oil Based on Wavelet Neural Network Optimized by Sparrow Search Algorithm

Abstract

Transformer oil is an indispensable part of the transformer. The performance of transformer oil largely determines the working condition of the transformer. Therefore, it is necessary to realize accurate and rapid detection of the physical and chemical properties of transformer oil. To realize the detection of transformer oil, firstly, the ultrasonic wave with multiple frequencies is transmitted through the transformer oil, and the physical data such as penetration velocity and attenuation coefficient in the transformer oil are detected. The parameters of the wavelet neural network were optimized using the sparrow search algorithm, and the feasibility of the model was detected using the transformer oil experimental data.