Assessment of Water Injection in a SI Engine using a fast running detailed chemistry based combustion model

Tim Franken (Brandenburg University of Technology Cottbus-Senftenberg), Lars Seidel (LOGE Deutschland GmbH), Andre Kulzer (Porsche AG)

CHALLENGE - Water injection is a key technology for turbocharged, spark ignition (SI) engines. It enables higher boost pressures and compression ratios, which results in increased engine efficiency. The rising number of optimization parameters pushes the traditional test bench approach to its limits and demands sophisticated simulation tools to support the engine development and precalibration.

SOLUTION - The 3D CFD test case for a boosted SI engine operating point at 2500 rpm and 16.2bar IMEP is used to perform the multi-objective optimization with the quasi-dimensional (QD) SI stochastic reactor model (SRM) with dual fuel (gasoline and water) tabulated chemistry. The multi-objective optimization process has been defined by using modeFRONTIER.  The Non-dominated Sorting Genetic Algorithm (NSGA-II) together with the Uniform Latin Hypercube (ULHC) space filler algorithm is applied. The number of individuals is set to 10 and the number of generations is 150. In total 1500 designs are calculated. The optimization target is to reduce the fuel mass and knock probability of the numerical test case.













BENEFITS - The general knowledge from SI engine development could be confirmed by the QD SRM optimization. The reduction of fuel consumption is favored by increased compression ratio and earlier spark timing. In contrast, the lower knock probability is influenced by low compression ratios and later spark timings. The presence of water is beneficial for reducing fuel consumption and knock probability at the same time. An optimum w/f ratio in the range of 25% to 30% is determined from the QD SRM simulation.