Virtual Engine Development: A Simulation Environment for Skip-Firing Strategy Optimization on IC Engines

The ever-increasing interest in global warming and air pollution has shaped the modern internal combustion engine. Over the years, new technologies were adopted to reduce emissions and fuel consumption. Various solutions have been implemented, such as Miller cycle, exhaust gas recirculation, multi-stage turbocharging, exhaust aftertreatment and cylinder deactivation. Wärtsilä has been investigating the latter technology for over a decade on its dual-fuel and natural gas engines, leading to a remarkable increase in engine efficiency at lower loads, as well as a more stable combustion process and lower hydrocarbon emissions. The aim of this master thesis project, conducted in collaboration with POWERTECH Engineering S.r.l. and Wärtsilä Italia S.p.a., is to propose and validate a methodology for the assessment, via 1D CAE simulation, of different skip-firing patterns in a large-bore medium-speed spark gas internal combustion engine. An engine performance model representing a Wärtsilä, 4-stroke, 16-cylinder gas engine, built in the commercial 1D-CFD simulation code GT-SUITE, was used as a baseline. A dedicated control logic was implemented, replicating the real-world engine skip-fire automation, in order to simulate different deactivation patterns and evaluate their impact on performance, consumption and hydrocarbon emissions. Simulation results were compared against corresponding experimental measurements, associated with different engine loads and varying deactivation patterns, obtaining an overall satisfactory correlation. The engine performance model was then coupled with a cranktrain mechanical model. Different integration techniques were explored to identify the most robust and reliable approach. An offline co-simulation process was eventually chosen and the impact of the different skip-fire pattern on torsional stresses and mechanical loads was evaluated, using the engine performance model to generate the boundary conditions required to run the cranktrain simulations. A streamlined and semi-automated process was therefore established, which enables the evaluation of different deactivation strategies and firing fractions in a single simulation environment. Results in terms of fuel efficiency increase potential, hydrocarbon emission reduction, along with stresses on the crankshaft, bearings and mechanical structure, can be evaluated predictively. This powerful simulation tool can be thus used to virtually optimize and evaluate a skip-fire strategy reducing the need for expensive and time-consuming experimental tests.