Method for feature extraction of hydroelectric generating set vibration fault

Abstract

The invention discloses a method for feature extraction of a hydroelectric generating set vibration fault. The method specifically includes the following steps: 1. using a Fast ICA(fast independent component analysis) method to process an original signal; 2. after Step 1, performing EEMD (ensemble empirical mode decomposition) on y in sequence; 3. after Step 2, calculating normalized energy and a normalized correlation coefficient of all IMFs (intrinsic mode functions) corresponding to y, and giving a corresponding energy threshold T and a system threshold epsilon; 4. after Step 3, selecting IMFs which satisfy requirements of the energy threshold T and the system threshold epsilon among x, taking a union set of the two as final effective IMFs, and performing reconstruction; and 5. after Step 4, finding out characteristic signals capable of representing set vibration from reconstruction signals. The method for feature extraction of a hydroelectric generating set vibration fault applies Fast ICA to feature extraction of hydroelectric generating set vibration signals, and solves the problem that a vibration fault in the hydroelectric generating set is difficult to accurately identify.

Description

technical field [0001] The invention belongs to the technical field of extraction methods, and relates to a feature extraction method for vibration faults of hydro-generator sets. Background technique [0002] The hydro-generator unit is very important in the hydropower plant, and its operation status is directly related to the safe and reliable operation of the whole system, so it is necessary to strengthen its condition monitoring and fault diagnosis. Vibration problems are very common in hydroelectric units, so they can be used as an important indicator to evaluate the operating status of the unit. However, due to the fact that various vibration faults may occur at the same time in the actual operation of the unit, these fault signals will affect and interfere with each other, and due to the existence of noise interference, the early and weak vibration signals will be completely submerged, and their waveforms will also appear distorted , so that the real fault informatio...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a feature extraction method for the vibration fault of a hydroelectric generating set, which combines fast independent component analysis and ensemble empirical mode decomposition and applies it to the feature extraction of the vibration signal of a hydroelectric generating set, solving the problem of hydroelectric power generation The problem that the vibration fault in the unit is difficult to be accurately identified

Images