Fault Diagnosis in Manufacturing Systems Using Machine Vision

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21220%2F21%3A00353844" target="_blank" >RIV/68407700:21220/21:00353844 - isvavai.cz</a>

Result on the web

<a href="https://doi.org/10.1109/PC52310.2021.9447495" target="_blank" >https://doi.org/10.1109/PC52310.2021.9447495</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/PC52310.2021.9447495" target="_blank" >10.1109/PC52310.2021.9447495</a>

Result language

angličtina

Original language name

Fault Diagnosis in Manufacturing Systems Using Machine Vision

Original language description

This paper describes the design and the verification of the method of fault diagnosis of production systems. A diagnosis system usually uses data obtained from a control loop (sensors, actuators, etc.). Unfortunately, clearly identify the fault is difficult or impossible. One set of symptoms can corresponds to more than one error. The conventional solution is to use additional sensors, however, this approach could lead to a parallel backup of the whole system. Here proposed solution is to use machine vision using OpenCV to obtain additional information about the state of the system, extend the set of symptoms and use them in the diagnostic module. In this work proposed diagnostic system is an extension of the previous designed one. A camera sensor was added. The data from the camera is evaluated and the diagnostic relevant objects are detected. These data are incorporated into the fault diagnostic system. The system was implemented and has been tested at a real simple electro-pneumatic circuit. The fault diagnostic system can detect hardware faults in the drive control loop like an electrical disconnection, low-pressure level, sensor, solenoid or PLC periphery malfunction.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2021

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 23rd International Conference on Process Control

ISBN

978-1-6654-0330-6

ISSN

—

e-ISSN

—

Number of pages

6

Pages from-to

313-318

Publisher name

IEEE

Place of publication

Piscataway

Event location

Štrbské Pleso

Event date

Jun 1, 2021

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

000723653400053