Author(s): Darian Onchis-Moaca | Gilbert-Rainer Gillich | Zeno-Iosif Praisach | Esperanza Maria Suarez-Sanchez
Journal: Analele Universităţii Eftimie Murgu Reşiţa. Fascicula de Inginerie
ISSN 1453-7397
Volume: XVII;
Issue: 1;
Start page: 251;
Date: 2010;
VIEW PDF
DOWNLOAD PDF
Original page
Keywords: intelligent agent | damage | identification | vibration | frequency
ABSTRACT
We introduce in this paper an intelligent system that automatically detects the damage in cantilever beams called AgentMec. This agent is creating its knowledge using a database with the first ten natural frequencies of the beam damaged at predetermined points. Using the reinforced learning, it creates the patterns to describe the behavior of the damaged beam. Then with each new vibration measurement send as input, AgentMec is able to recognize the location of the damage.
Journal: Analele Universităţii Eftimie Murgu Reşiţa. Fascicula de Inginerie
ISSN 1453-7397
Volume: XVII;
Issue: 1;
Start page: 251;
Date: 2010;
VIEW PDF


Keywords: intelligent agent | damage | identification | vibration | frequency
ABSTRACT
We introduce in this paper an intelligent system that automatically detects the damage in cantilever beams called AgentMec. This agent is creating its knowledge using a database with the first ten natural frequencies of the beam damaged at predetermined points. Using the reinforced learning, it creates the patterns to describe the behavior of the damaged beam. Then with each new vibration measurement send as input, AgentMec is able to recognize the location of the damage.