The Oil & Gas industry is on the threshold of digital transformation through integrating Digital Twins and Artificial Intelligence. However, the widespread adoption of this technology is still limited. This study introduces an innovative use of Digital Twins based on models obtained through artificial intelligence to analyse a vertical tank behaviour of an experimental plant. Moving beyond the traditional non-real-time analysis software that currently dominates plant operations, this approach leverages real-time data to advance the modelling process. From the previous research about using artificial neural networks to model an ejector, the present work expands the scope to include the vertical reservoir. It adds a new piece to constructing a system that can correctly describe the experimental plant and detect its anomalies. The tank model is realised through two artificial intelligence algorithms that accurately predict pressures and water levels inside the tank at the “t+1” time step. These algorithms have been rigorously trained and tested with real plant data, demonstrating high fidelity in modelling tank behaviour with an accuracy of 99.98% and 99.75%. With this experimental case, the synergy between Artificial Intelligence and Digital Twins demonstrates its relevance in real-time Oil & Gas plant management. It underscores the potential for transformation to enable more dynamic, resilient, effective and safe plant operations

Artificial Intelligence and Digital Twin for Effective Risk Management in an experimental plant / Pietrangeli, I.; Mazzuto, G.; Ciarapica, F. E.; Bevilacqua, M.. - 2024-September:(2024). (Intervento presentato al convegno 36th European Modeling and Simulation Symposium, EMSS 2024, Held at the 21st International Multidisciplinary Modeling and Simulation Multiconference, I3M 2024 tenutosi a Tenerife, Spain nel 18 - 20 September 2024) [10.46354/i3m.2024.emss.009].

Artificial Intelligence and Digital Twin for Effective Risk Management in an experimental plant

Pietrangeli I.
;
Mazzuto G.;Ciarapica F. E.;Bevilacqua M.
2024-01-01

Abstract

The Oil & Gas industry is on the threshold of digital transformation through integrating Digital Twins and Artificial Intelligence. However, the widespread adoption of this technology is still limited. This study introduces an innovative use of Digital Twins based on models obtained through artificial intelligence to analyse a vertical tank behaviour of an experimental plant. Moving beyond the traditional non-real-time analysis software that currently dominates plant operations, this approach leverages real-time data to advance the modelling process. From the previous research about using artificial neural networks to model an ejector, the present work expands the scope to include the vertical reservoir. It adds a new piece to constructing a system that can correctly describe the experimental plant and detect its anomalies. The tank model is realised through two artificial intelligence algorithms that accurately predict pressures and water levels inside the tank at the “t+1” time step. These algorithms have been rigorously trained and tested with real plant data, demonstrating high fidelity in modelling tank behaviour with an accuracy of 99.98% and 99.75%. With this experimental case, the synergy between Artificial Intelligence and Digital Twins demonstrates its relevance in real-time Oil & Gas plant management. It underscores the potential for transformation to enable more dynamic, resilient, effective and safe plant operations
2024
979-12-81988-02-6
File in questo prodotto:
File Dimensione Formato  
Pietrangeli_Artificial-intelligence-digital-twin_2024.pdf

accesso aperto

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza d'uso: Creative commons
Dimensione 1.06 MB
Formato Adobe PDF
1.06 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/11566/341093
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact