Development of a phenomenological spray model for next generation dual-fuel marine engines: validation in free-field conditions

Abstract

This study develops and validates a reduced-order phenomenological spray model for dual-fuel marine engines operating under reactivity controlled compression ignition (RCCI) conditions. Rather than introducing a new spray-theory class, the work adapts a Musculus-based formulation into an RCCI-oriented framework capable of predicting both spray penetration and cylinder-fixed fuel–air stratification at low computational cost. The main methodological advance is the introduction of a CV-to-zone mapping strategy for transferring spray control-volume information into the zonal structure required by multi-zone combustion models; among the three tested approaches, an angled non-uniform mapping gave the most accurate stratification representation. Validation is via high-speed optical measurements by the Optical Spray Combustion Chamber and high-fidelity CFD simulations with CONVERGE. Sensitivity analysis identified the discharge coefficient (Ca) and velocity profile factor (β) as dominant calibration parameters, with profile evolution factor (α) serving as a secondary shaper of radial distribution. A two-stage optimization strategy for local tuning per case and global tuning across conditions demonstrated that the globally calibrated model reproduces spray penetration within ∼14% of CFD, and peak high-reactivity fuel location within ±2 zones across calibration and validation datasets. Amplitude prediction of peak fuel fraction exhibited larger scatter (∼40%), but location fidelity was robust. The framework provides a reliable and computationally efficient surrogate for spray-mixing dynamics and is well suited for future integration with fast 0D/1D multi-zone combustion models to support predictive, cycle-resolved simulation of advanced RCCI engines.