The Transformer Winding Deformation Detection System Based On Ultrasonic Ranging Method

Winding deformation is one of the hidden dangers of the safe operation of transformer.However,the existing winding deformation detection methods cannot achieve on-line detection,and cannot judgment fault location intuitively.Therefore,an ultrasonic detection method based on the distance transformation algorithm is proposed and the transformer winding deformation intelligent detection system based on ultrasonic ranging is designed in this paper.The intuitive,quantitative charged detection of the transformer winding is implemented.First of all,the basic principle of transformer winding deformation using ultrasonic detection is introduced in this article,and the ultrasonic probe placement algorithm is proposed to improve the detection range of ultrasonic method.In order to improve the effect of the ultrasonic signal reception,four different cases of the ultrasonic probe placement is studied in this algorithm.A distance transform algorithm based on reflection loop is proposed in this paper to get the distance between the test point and the transformer case.Secondly,the transformer winding deformation detection hardware system based on ultrasonic ranging is designed in this paper.The electrical contact between detection device and transformer is avoided and the charged detection is implemented.The suitable ultrasonic transducer is selected by calculating the ultrasound effective frequency in transformer oil.The ultrasonic wave refraction loss is minimized,the ultrasonic propagation distance in transformer oil is increased and the measurement accuracy of the ultrasonic time signal is guaranteed through the above work.Finally,a transformer winding state three-dimensional imaging software using Matlab GUI is developed in this paper.This software can generate real-time three-dimensional images of transformer winding at the detection scene,and provide analytical tools such as transformer deformation contrast figure.The intuitive,quantitative charged detection of the transformer winding is implemented.