基于压缩感知的设备多源信息传输与分类算法

Translated title of the contribution: Multi-source information transmission and classification algorithm for equipment based on compressed sensing

Xiaohu Zhao, Gang Wang, Boming Song, Jiacheng Yu

Research output: Contribution to journalArticle

Abstract

Aiming at the characteristics of various types of equipment in coal preparation plant and the dispersion of monitoring points, a multi-source information wireless transmission and classification algorithm for equipment based on compressed sensing was proposed. By constructing a multi-hop information transmission model, the information transmission problem was transformed into the compressed sensing problem of multi-path measurement signals, thereby the measurement matrix acquisition was transformed into the routing problem of the multi-hop information transmission model. Aiming at the large coherence of the obtained measurement matrix and affecting the signal reconstruction effect, the idea of random routing was introduced into the routing construction, and a random dynamic self-organizing routing algorithm was proposed. In order to solve the problem that the time domain features of the reconstructed signal were difficult to accurately classify the fault type, a new time domain feature, the total variation (TV) of the vibration signal, was introduced for the reconstructed signal, and the compensation distance estimation algorithm was adopted to verify the superiority of the introduction of indicators. The analysis of the measured data of the coal preparation plant shows that the proposed multi-source information transmission and classification algorithm can effectively improve the fault recognition accuracy under the condition of improving the real-time transmission efficiency of the monitoring data.

Original languageChinese
Pages (from-to)13-24
Number of pages12
JournalTongxin Xuebao/Journal on Communications
Volume41
Issue number2
DOIs
Publication statusPublished - 25 Feb 2020

Fingerprint Dive into the research topics of 'Multi-source information transmission and classification algorithm for equipment based on compressed sensing'. Together they form a unique fingerprint.

  • Cite this