Partial power discharge fault determining characteristics extracting method and determining method

Abstract

The invention relates to the monitoring technology field and more particularly, to a partial power discharge fault determining characteristics extracting method and a determining method. The invention provides a partial power discharge fault type determining characteristics extracting method, comprising: carrying out discrete sampling to the signals from a partial power discharge area; building a Hankel matrix on the discrete digital signals; performing singular value decomposition to the Hankel matrix to obtain the singular value matrixes; screening the singular values with the preserving of the larger singular values to obtain the screened singular value matrixes; based on the screened singular value matrixes, obtaining the improved Hankel matrix; performing phase-space reconstruction to the improved Hankel matrix to obtain the constructed phase-space matrix B; and calculating the energy moments of all the components of the phase-space matrix B to obtain the partial power discharge fault determining characteristic vector. With the method of the invention, it is possible to effectively avoid the problem that the diagnosing precision is influenced by the unequal number of the valid singular values generated from the singular value decomposition so as to improve other methods which neglect the mutual inherent relationship of the characteristic values in mode identification.

Description

technical field [0001] The invention relates to the technical field of monitoring, in particular to a partial discharge fault judgment feature extraction method and a judgment method. Background technique [0002] Partial discharge is the main cause of failure and insulation deterioration of high-voltage electrical equipment, and it is also the main symptom and manifestation of insulation deterioration. Therefore, a great deal of research on partial discharge has been done in monitoring the insulation condition of large electric equipment. In terms of partial discharge fault diagnosis, being able to effectively extract the fault characteristics of partial discharge, and accurately realize on-line monitoring and real-time fault diagnosis will provide an important guarantee for the safe and stable operation of the power system. [0003] At present, when dealing with partial discharge signals, the commonly used feature extraction analysis methods include statistical parameter ...

AI Technical Summary

Problems solved by technology

The singular value decomposition method can detect mutation information in the background of strong noise, and shows unique advantages in signal processing. It has the advantages of zero phase shift, small waveform distortion, and high signal-to-noise ratio, but the singular value decomposition has different extraction states. The problem that the number of feature components obtained from information is not equal

Images