Ultrasonic Signal Path Identification and Search-Based Localization of Partial Discharges in Power Transformers

The intricate internal structure of power transformers causes ultrasonic waves generated by partial discharge (PD) sources to change direction and velocity as they propagate. Accurate localization of PD sources requires accounting for these changes rather than assuming a straight-line path from the source to the sensor. To ensure high-precision localization even when the PD source is far from the ultrasonic sensors, this paper proposes a PD localization algorithm based on ultrasonic wave propagation paths. First, the influence of windings, the iron core, and the tank wall on ultrasonic wave direction and velocity is analyzed through acoustic theory and simulation. A transformer simulation model is then developed, and an optimization method for PD localization is proposed. The key advantage of this method is that it considers the propagation path’s effect on time-of-flight (TOF) during the localization process, offering a more realistic reflection of actual ultrasonic wave behavior. To validate the effectiveness of the proposed simulation method and its impact on localization accuracy, experiments were conducted on a 110 kV power transformer, and ultrasonic signals from internal PD were collected. Experimental results show that the average localization error using the least squares method is 38.83 cm, whereas the proposed method achieves an average error of only 13.43 cm—reducing the error by 25.4 cm. This confirms that the proposed method effectively improves the accuracy of PD localization in power transformers.