Transformer Defect Location Based on Distributed Oil Chromatography Analysis and Improved Time Delay Estimation

In response to the challenges existing in transformer partial discharge positioning, this paper proposes an innovative comprehensive positioning technology that combines distributed oil chromatography analysis with an improved time delay estimation method. First, distributed oil chromatography monitoring is used to roughly locate the location of potential defects, and the changes in acetylene content at different oil intake ports of the transformer are used to identify the general area of the partial discharge source. Subsequently, the ultrasonic sensor is constructed as a four-element cross array and an improved generalized cross-correlation delay estimation procedure is employed. The improved extended mutual dependence delay estimation algorithm includes the following key steps: First, grounded in the traditional extended mutual dependence, the quadratic cross-correlation technology is introduced to further enhance the correlation derived from the waveform, consequently elevating the robustness of the lag determination. and accuracy; then, in the peak detection stage, the cubic spline interpolation method is used to refine the highest magnitude from the mutual dependence relation. Spline interpolation can smooth the data within a local range, improve the accuracy of peak detection, and thus estimate more accurately. Delay to improve delay estimation accuracy in noisy environments.Experiments have shown that the proposed method is superior to the traditional algorithm in terms of the deviation of the sound source coordinates. This method can not only effectively suppress noise interference, but also significantly enhance the location precision of the localized emission waveform.