Health assessment framework of marine engines enabled by digital twins

The advancements in digital twins when combined with the use of machine learning tools can facilitate effective health assessment and diagnostics safety critical systems. This study aims at developing a framework to address marine engines utilising based on first-principles. follows four distinct stages, former two including engine digital-twin set up by customising required thermodynamic models, as well its calibration using tests trials data representing healthy conditions. In third stage, measurements from actual operating conditions are corrected subsequently employed develop twin prevailing fourth stage deals assessing metrics derived developed twins. is demonstrated case large four-stroke nine-cylinder propulsion engine. results demonstrate that three cylinders identified be underperforming leading an average increase Brake Specific Fuel Consumption (BSFC) 2.1%, whereas decreases 6.8% Indicated Mean Effective Pressure (IMEP) 6.1% Exhaust Gas Temperature (EGT) exhibited for across entire envelope. mapping performance envelope under several conditions, providing enhanced insights current status. advantages proposed include easily obtained data, application types both auxiliary use.