This research paper focuses on the utilization of Virtual Gas Analyzers (VGAs) within the maritime industry to observe, analyze and predict gas compositions in marine holds. The maritime sector encounters numerous challenges concerning the efficient transportation, particularly considering increasingly stringent regulations. By enabling real-time predicting of water flow, VGAs offer a solution to prevent environmental pollution. The objective was to collect relevant data from IoT sensorts and present it in a comprehensive Machine Learning (ML) model to operators responsible for on board vessels. It works on modeling for exhaust gas cleaning systems (EGCS) and assessing the impact of wash water discharges. The study reveals research on VGAs prediction models which regarding their usage of ML models such as Linear Regression, Random Forest, LightGBM and Multi-Layer Perceptron. The prediction model that VGAs hold great promise in enhancing efficiency and accuracy are capable of effective prediction. Added to the model, the criteria need to consider is how the systems will behave in different situations which, handled by Polynomial chaos expansion (PCE). The study concludes with recommendations for policies and practices, underscoring the significance of ongoing innovation and investment in this field.

Virtual Gas Analyzer in the Maritime Sector

Bruzzone Agostino;De Paoli, Alberto;
2023-01-01

Abstract

This research paper focuses on the utilization of Virtual Gas Analyzers (VGAs) within the maritime industry to observe, analyze and predict gas compositions in marine holds. The maritime sector encounters numerous challenges concerning the efficient transportation, particularly considering increasingly stringent regulations. By enabling real-time predicting of water flow, VGAs offer a solution to prevent environmental pollution. The objective was to collect relevant data from IoT sensorts and present it in a comprehensive Machine Learning (ML) model to operators responsible for on board vessels. It works on modeling for exhaust gas cleaning systems (EGCS) and assessing the impact of wash water discharges. The study reveals research on VGAs prediction models which regarding their usage of ML models such as Linear Regression, Random Forest, LightGBM and Multi-Layer Perceptron. The prediction model that VGAs hold great promise in enhancing efficiency and accuracy are capable of effective prediction. Added to the model, the criteria need to consider is how the systems will behave in different situations which, handled by Polynomial chaos expansion (PCE). The study concludes with recommendations for policies and practices, underscoring the significance of ongoing innovation and investment in this field.
File in questo prodotto:
File Dimensione Formato  
hms_2023_013.pdf

accesso aperto

Tipologia: Documento in Post-print
Dimensione 832.17 kB
Formato Adobe PDF
832.17 kB 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/11567/1160234
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact