Ultrasonic echo denoising method for ultrasonic testing of transformer high voltage bushing lead

Abstract

The invention relates to an ultrasonic echo denoising method for ultrasonic testing of a transformer high voltage bushing lead. The method comprises the steps of: (1) performing wavelet decompositionon a ultrasonic echo signal reflected by a transformer high voltage bushing lead; (2) performing denoising preprocessing on the ultrasonic echo signal; (3) performing EEMD decomposition on the denoising preprocessed echo signal to obtain a multi-layer IMF component; 4) solving the noise component standard deviation of the denoising preprocessed signal, wherein the first two layers of the IMF component filter the noise in the signal by using a new method "2 sigma rule", and remaining IMF component denoise using an adaptive threshold; (5) calculating a correlation coefficient between the twice denoised IMF component and original component, selecting the IMF component to be retained according to the coefficient for a signal reconstruction, and obtaining the denoised ultrasonic echo signal. Compared with the prior art, the method can effectively filter out random noise in the ultrasonic echo signal, maximize the integrity of the echo signal, and better recognize the oscillation starting position of the echo signal.

Description

technical field [0001] The invention relates to the field of ultrasonic detection of transformer high-voltage bushing lead wires, in particular to an ultrasonic echo denoising method for ultrasonic detection of transformer high-voltage bushing lead wires. Background technique [0002] As the key equipment of the power system, the power transformer is closely related to the safety and stability of the power system. The high-voltage bushing is an important part of the power transformer. The lead wire of the transformer winding must pass through the insulating sleeve to connect to the external circuit. The high voltage bushing is the insulation between the lead wires and between the lead wire and the oil tank, and plays the role of fixing the lead wire. During the operation of the equipment, the high-voltage bushing has to bear the effects of various factors such as heat, electricity, chemistry, and machinery for a long time, resulting in the deterioration of the insulation and...

AI Technical Summary

Problems solved by technology

[0004] At present, in order to reduce the influence of noise, the hardware considers to use the filter to denoise when the signal is collected, but this method can only filter out the noise signal with a large amplitude, and the denoising effect is not thorough; currently the widely used ultrasonic echo Wavelet denoising techniques mainly include wavelet denoising, EMD (empirical mode decomposition) denoising, EEMD (ensemble empirical mode) denoising and other methods.

Wavelet denoising technology has good flexibility, decorrelation and other characteristics, but there are problems such as difficulty in wavelet base selection, difficulty in determining decomposition levels, inaccurate threshold values, and poor adaptability.

Compared with wavelet denoising technology, EMD denoising technology overcomes the shortcomings of poor adaptability in the signal decomposition process, and does not need any basis function, so the denoising effect is better, but it still has algorithm defects such as modal aliasing and endpoint effect

In view of the defects of EMD denoising technology, an EEMD denoising technology has been proposed in recent years. This method introduces Gaussian white noise on the basis of EMD decomposition, so that the continuity of the noise-contaminated signal on different frequency scales is improved, and the denoising The effect has been further improved, and the problem of modal aliasing has been suppressed, but there are problems such as large amount of calculation and susceptibility to interference

Images