Login / Signup

Design of Evaluation Classification Algorithm for Identifying Conveyor Belt Mistracking in a Continuous Transport System's Digital Twin.

Gabriel FedorkoVieroslav MolnárBeata StehlikovaPeter MichalikJan Saliga
Published in: Sensors (Basel, Switzerland) (2024)
A prerequisite for continuous transport systems' operation is their digital transformation, which interprets operating conditions based on the availability of a wide range of data and information in the form of measured quantities that can be obtained, for example, by experimental measurement. To implement digital transformation in continuous transport systems, it is necessary to examine and analyze the informative value of individual measured quantities in detail. Research in this area must focus on identifying addressable quantities with a clear, informative value. Such an approach enables the monitoring of continuous transport systems operation and performance of operational diagnostics, the objective of which should be identifying undesirable operating conditions. Within this paper, research will be presented aiming to verify the hypothesis that, based on a measurement of selected parameters, it is possible to identify belt mistracking in a continuous transport system. Belt mistracking is an undesirable condition that can cause a conveyor belt to converge and thus seriously turn off an entire transport system. The research results confirmed the established hypothesis. Based on this, an evaluation algorithm was created for on-time evaluation. The proposed algorithm is also suitable for the needs of a digital twin of a continuous transport system.
Keyphrases
  • machine learning
  • deep learning
  • healthcare
  • neural network
  • social media
  • quantum dots