logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-01282022-223134


Tipo di tesi
Tesi di laurea magistrale
Autore
FONTANELLI, LEONARDO
URN
etd-01282022-223134
Titolo
Development of a predictive model for production data based on machine learning
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
COMPUTER ENGINEERING
Relatori
relatore Prof. Cimino, Mario Giovanni Cosimo Antonio
correlatore Prof.ssa Vaglini, Gigliola
tutor Dott. Parlangeli, Gianluca
Parole chiave
  • anomalies
  • data analysis
  • data engineering
  • industrial
  • machine learning
  • vibrations
Data inizio appello
18/02/2022
Consultabilità
Non consultabile
Data di rilascio
18/02/2062
Riassunto
The thesis work is the result of a company internship at Extra Red and the project had the aim of designing, developing and deploy a machine learning model for an industrial winding machine.
The machine learning model will allow to predict anomalous behavior of an industrial machine and also to exploit the reasons that causes those anomalies. The problem have been addressed by following a machine learning process pipeline and data analysis techniques. The work starts with an understanding of the legacy project and then gathering and engineering the data of the production machine for a coherent dataset. Hence, with the cleaned and engineered dataset, it have been possible to train machine learning models for predicting and also to understand the vibration phenomenon in real time production of the industrial machine.
File