Application of Modern Approaches to Solving Diagnostic Tasks in Industry

článek ve sborníku ve WoS nebo Scopus

en

Technical diagnostics is a separate scientific discipline focused on non-destructive methods of assessing the technical condition of objects and is one of the pillars of reliability and functionality of equipment. The definitions of technical diagnostics are based on various disciplines and standards, and technical diagnostics relies on the monitoring of diagnostic variables such as temperature, vibration, noise, and others. The paper further discusses the types of diagnostics, including vibrodiagnostics, thermodiagnostics, electrodiagnostics, etc., and their division into on-line and offline systems. The next section is focused on data evaluation, which is based on standards and methods, including frequency analysis and the use of artificial intelligence. The paper also discusses the use of cloud computing, edge computing, and risk theory, which can contribute significantly to the efficiency of diagnostic processes in the future. Finally, practical examples that have been solved at our department are presented. The first one is the use of the SIPLUS CMS1200 system for on-line vibration monitoring, which demonstrates its functionality and possibilities of use in industrial practice. The second example is the possibility of using neural networks in the processing and evaluation of measured data. The last example focuses on the application of risk theory in the evaluation of different diagnostic methods. The article aims to introduce the reader to the issues of technical diagnostics and then demonstrate possible approaches for practical solutions.

Technical diagnostics is a separate scientific discipline focused on non-destructive methods of assessing the technical condition of objects and is one of the pillars of reliability and functionality of equipment. The definitions of technical diagnostics are based on various disciplines and standards, and technical diagnostics relies on the monitoring of diagnostic variables such as temperature, vibration, noise, and others. The paper further discusses the types of diagnostics, including vibrodiagnostics, thermodiagnostics, electrodiagnostics, etc., and their division into on-line and offline systems. The next section is focused on data evaluation, which is based on standards and methods, including frequency analysis and the use of artificial intelligence. The paper also discusses the use of cloud computing, edge computing, and risk theory, which can contribute significantly to the efficiency of diagnostic processes in the future. Finally, practical examples that have been solved at our department are presented. The first one is the use of the SIPLUS CMS1200 system for on-line vibration monitoring, which demonstrates its functionality and possibilities of use in industrial practice. The second example is the possibility of using neural networks in the processing and evaluation of measured data. The last example focuses on the application of risk theory in the evaluation of different diagnostic methods. The article aims to introduce the reader to the issues of technical diagnostics and then demonstrate possible approaches for practical solutions.

technical diagnostics, neural networks, vibrodiagnostics, condition monitoring, risks

06.12.2024

IEEE

Brno, Czech Republic

979-8-3503-9490-0

2024 21st International Conference on Mechatronics - Mechatronika (ME)

1

7