Modelling of auto-ignition to flame propagation transition for Dual-Fuel combustion

Dual-Fuel (DF) combustion engines are a promising alternative to conventional technologies in terms of efficiency and emissions in particular for heavy duty vehicles or marine engines. DF engines involve the injection of a high reactivity pilot spray as a source of ignition for a lean and low reactivity premixed main charge. As a result, auto-ignition (AI) spots emerging from pilot spray generate premixed flame (PF) combustion of the main charge. Nevertheless, transition from AI to PF is not fully understood and thus challenging in terms of modelling. The main goal of this work is to propose a predictive model valid for both AI and PF regime, including transition phenomena, in a Large-Eddy Simulation (LES) context. Firstly, a set of one-dimensional study is conducted. The limitations of two classical approaches for describing auto-ignition (AI), premixed flame (PF) propagation, and regime transitions are identified. The analysis reveals that transition from AI to PF is rather progressive, and starts with hot flame ignition (HFI). Thickened Flame Model (TFM), including detailed resolution of chemistry, and reveals that standard formulation is unable to represent AI, as thickening factor dramatically increases AI delays. A modelling approach involving a thickening factor relaxation time is proposed for LES of DF engines applications, and appears capable of representing transition from AI to PF phenomena in reactivity stratified situations typically encountered in DF applications. The potential of this approach, its accuracy and its limitations are assessed by comparing the model behaviour against 1D reference test cases, and measurements on an optical test bench.