The Industry 4.0 approach represents the driving force behind the current industrial research and development activities. The spread of smart sensors and actuators, tightly connected in the Industrial Internet of Things, offers a wide range of possibilities for the application of control methodologies to tackle very different control problems. In addition, the huge amount of data captured by sensors can be exploited to develop and validate very accurate models of the processes under control, thus allowing to perform preliminary studies in a simulated environment and move to the real plant only when the control approach is sufficiently consolidated. In this framework, this Thesis faces three different control problems with the application of various control methodologies. In particular, a Proportional-Intergral-Derivative is adopted to perform fine regulation of magnetic field in Fast Field Cycling Nuclear Magnetic Resonance experiments, with a closed-loop based on an ad-hoc virtual sensor for magnetic field measurements. Both the design of the virtual sensor and the synthesis of the regulator are studied in simulation first, then successfully applied to the real plant. The second task faced in this work is service pressure regulation in Water Distribution Networks. Different control algorithms (frequency domain and optimal control) are developed and tested in a simulated environment to investigate how to take advantage of the improved communication capabilities made available by wired networks, where pressure sensors and control valves are connected by wire to a digital control unit. Finally, the last application faced in this Thesis is related to smart warehousing, and deals with the control of mechanical systems with actuators affected by deadzone nonlinerity. In particular, two different Model Predictive Control approaches are tested both in simulation and on a laboratory scale version of an overhead travelling crane, with the aim of assessing their performances in term of accuracy, robustness and computational honour. The results obtained in the above case studies stresses how the application of consolidated control methodologies, available from control theory, allows to successfully face the new challenges that arise by the Industry 4.0 context.

Control Applications in the Context of Industry 4.0

GALUPPINI, GIACOMO
2020-02-26

Abstract

The Industry 4.0 approach represents the driving force behind the current industrial research and development activities. The spread of smart sensors and actuators, tightly connected in the Industrial Internet of Things, offers a wide range of possibilities for the application of control methodologies to tackle very different control problems. In addition, the huge amount of data captured by sensors can be exploited to develop and validate very accurate models of the processes under control, thus allowing to perform preliminary studies in a simulated environment and move to the real plant only when the control approach is sufficiently consolidated. In this framework, this Thesis faces three different control problems with the application of various control methodologies. In particular, a Proportional-Intergral-Derivative is adopted to perform fine regulation of magnetic field in Fast Field Cycling Nuclear Magnetic Resonance experiments, with a closed-loop based on an ad-hoc virtual sensor for magnetic field measurements. Both the design of the virtual sensor and the synthesis of the regulator are studied in simulation first, then successfully applied to the real plant. The second task faced in this work is service pressure regulation in Water Distribution Networks. Different control algorithms (frequency domain and optimal control) are developed and tested in a simulated environment to investigate how to take advantage of the improved communication capabilities made available by wired networks, where pressure sensors and control valves are connected by wire to a digital control unit. Finally, the last application faced in this Thesis is related to smart warehousing, and deals with the control of mechanical systems with actuators affected by deadzone nonlinerity. In particular, two different Model Predictive Control approaches are tested both in simulation and on a laboratory scale version of an overhead travelling crane, with the aim of assessing their performances in term of accuracy, robustness and computational honour. The results obtained in the above case studies stresses how the application of consolidated control methodologies, available from control theory, allows to successfully face the new challenges that arise by the Industry 4.0 context.
26-feb-2020
File in questo prodotto:
File Dimensione Formato  
PhDThesis_Galuppini_Final.pdf

accesso aperto

Descrizione: tesi di dottorato
Dimensione 28.92 MB
Formato Adobe PDF
28.92 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11571/1322867
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact