An amplitude-frequency time-varying process signal segmentation method based on a sliding window

Abstract

The invention discloses an amplitude-frequency time-varying process signal segmentation method based on a sliding window. The method comprises the following steps of: (1) setting a sliding window foran amplitude-frequency time-varying process signal to be analyzed; (2) calculating a mean value of each process signal point in the sliding window, and dynamically setting a segmentation threshold point under the process curve according to the mean value so as to obtain a lower segmentation line; (3) determining each lower segmentation point of the process curve according to the lower segmentationlines, and segmenting the process curve according to the lower segmentation points; And (4) calculating the similarity of the segmentation curves by using the process signals in the segmentation curves, and carrying out segmentation effectiveness judgment. According to the method, process period segmentation is carried out on the amplitude-frequency time-varying process curve, and the method canadapt to the change of process parameters and has relatively high adaptability, stability and robustness.

Description

technical field [0001] The invention relates to the technical field of process signal segmentation, in particular to an amplitude-frequency time-varying process signal segmentation method based on a sliding window. Background technique [0002] In the field of fault diagnosis and equipment status monitoring, both rely on the extraction of the characteristic value of the process curve, and one way of feature value extraction is to segment the process curve, and then perform feature extraction on the segmented curve. Therefore, the process curve segmentation algorithm is the basis of equipment fault diagnosis and operation status monitoring. [0003] At present, there are many researches on the segmentation algorithm of equipment operation process curve. The general method is: first define the division of functional segments in the signal cycle according to the extreme point of the signal, monotonicity or the actual meaning of the signal, and then use the time between signal s...

AI Technical Summary

Problems solved by technology

[0004] However, in actual production and processing, the process parameters change in real time, resulting in the equipment operation process curve is a variable process parameter curve, the amplitude and frequency of the process curve will change with the change of process parameters, the existing signal segmentation method cannot Errors often occur when adapting to process signals whose periods and extreme values ​​change

Images