A Multi-source Signal Separation Method Based on Non-negative Tensor Decomposition

A non-negative tensor decomposition, source signal technology, applied in measurement devices, measuring ultrasonic/sonic/infrasonic waves, instruments, etc., can solve the problem of reducing the decomposition effect of the source signal, unable to adapt to different vibration signal characteristics, unable to fully characterize the source signal vibration features, etc.

Active Publication Date: 2020-06-26
XI AN JIAOTONG UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The existing non-negative tensor is mainly constructed by combining two-dimensional time-frequency data with channels, while the window length in the time-frequency transformation is selected based on experience, which cannot adapt to different vibration signal characteristics, and in the process of non-negative tensor decomposition The number of source signals is also given by experience, which leads to the inability to fully characterize the vibration characteristics of the source signal and reduces the decomposition effect of the source signal

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Multi-source Signal Separation Method Based on Non-negative Tensor Decomposition
  • A Multi-source Signal Separation Method Based on Non-negative Tensor Decomposition
  • A Multi-source Signal Separation Method Based on Non-negative Tensor Decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.

[0032] refer to figure 1 , a multi-source signal separation method based on non-negative tensor decomposition, including the following steps:

[0033] Step 1: Generate four typical analog signals, namely, FM-AM signal, AM signal, AM-FM signal and periodic impact signal. Among them, the FM-AM signal simulates the meshing vibration of gears, and the carrier frequency and modulation frequency are respectively 750Hz and 25Hz. The carrier frequency and modulation frequency of the AM signal are 400Hz and 10Hz respectively, the carrier frequency and modulation frequency of the AMFM signal are 210Hz and 60Hz respectively, the oscillation frequency of the periodic impact signal is 3000Hz, and 2000 data points of each signal are taken. Get the multi-source vibration signal A 2000×4 , whose waveform and spectrum refer to figure 2 and image 3 shown;

[003...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A multi-source signal separation method based on non-negative tensor decomposition. For multi-source vibration signal data, the optimal window length is determined based on the average information entropy method, and then the short-time Fourier transform is adopted on the basis of the optimal window length Construct a three-dimensional time-frequency tensor distribution, and use the number of iteration steps, convergence error and kernel consistency index to select the optimal number of source signals, perform non-negative tensor decomposition according to the optimal number of source signals, and use the matrix weight obtained by the decomposition The time-frequency matrix distribution of the source signal is constructed, and then the source signal is obtained by inverse short-time Fourier transform. The invention can accurately extract the source signal from the mixed signal.

Description

technical field [0001] The invention belongs to the technical field of equipment detection and fault diagnosis, and in particular relates to a multi-source signal separation method based on non-negative tensor decomposition. Background technique [0002] With the continuous improvement of the complexity of mechanical equipment, the collected vibration signals are often a mixture of multiple vibration excitation source components, and under the interference of noise, it is difficult to identify the characteristic components of the faulty components, which affects the identification of faults. Therefore, how to separate the main characteristic components from the vibration signal is one of the urgent problems in fault diagnosis. As a new direction in the field of modern signal processing, the signal source separation technology separates the source signal from the obtained mixed signal when the number, location and transmission channel of the source signal are unknown. The co...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G01H17/00
CPCG01H17/00G06F18/2133G06F18/21347
Inventor 刘弹李光梁霖刘飞王宝栗茂林
Owner XI AN JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap