A Joint Work to Develop a Predictive 1D Modelling Approach for Heavy Duty Gaseous Fueled Engines through Experiments and 3D CFD Simulations

The present paper reports experimental and numerical research activities devoted to deeply characterize the behavior and performance of a Heavy Duty (HD) internal combustion engine fed by compressed natural gas (CNG). Current research interest in HD engines fed by gaseous fuels with low C/H ratios is related to the well-known potential of such fuels in reducing carbon dioxide emissions, combined to extremely low particulate matter emissions too. Moreover, methane, the main CNG component, can be produced through alternative processes relying on renewable sources, or in the next future replaced by methane/H2 blends. The final goal of the presented investigations is the development of a predictive 0D combustion submodel within the framework of a 1D numerical simulation platform. To this aim, an experimental campaign has been carried out on a six-cylinder HD spark ignition engine CNG engine, Euro VI d compliant, typically employed in road vehicle applications, at the test bench, in order to build a comprehensive and extended database. The experimental characterization was necessary not only to have a defined picture of the engine behavior, but also to provide the required initial and boundary conditions and a consistent dataset for 1D and 3D models validation. Then, full-cycle 3D CFD numerical simulations have been carried out, reproducing all the engine phases of a selected cylinder: it has thus been possible to further enrich the set of information regarding main fluid-dynamic features of the investigated geometry and corresponding combustion evolution. At the same time, a 1D model of the full engine layout has been built. At first, it was preliminary calibrated and validated through a non-predictive combustion submodel (Three Pressure Analysis approach). Finally, relying on experimental and predicted data, including global swirl ratio temporal evolution, turbulent intensity and length scale, it has been possible to set up a predictive modelling approach, capable of suitably reproducing pressure profiles and flow rates in various engine operating conditions.