Visual multi-objective optimization of the performance of a two-stroke aviation piston engine with the predictive combustion model at different altitude based on 3D scavenging computation

To enable more effective design and analysis of aviation piston engine performance at different altitudes, a two-stroke aviation piston engine was selected for simulation investigation. Unlike four-stroke engines, two-stroke engines have distinct scavenging characteristics, and traditional 1D simulation models are often not accurate in their combustion prediction. To address this issue, a 1D simulation approach was developed by integrating 3D scavenging computation. First, the accuracy of the model was validated using experimental data of different ignition timings. Subsequently, a comparative analysis of combustion performance between the predictive SI-Turb combustion model and the non-predictive SI-Wiebe model was conducted. Then, the variation of scavenging characteristics at different conditions was considered, and engine performance was optimized across multiple altitudes. Three optimization objectives were defined, including engine brake power, peak cylinder pressure, and brake specific fuel consumption (BSFC). Ignition timing, fuel injection timing, and excess air coefficient are factors in the optimization. The Latin Hypercube sampling design was used to generate simulation cases. The response surface method (RSM) was used to establish the correlations between the factors and the optimization objectives at different altitudes. After the response surface models were validated in accuracy, a new two-dimensional optimization (2DO) method was used to generate optimization contour maps. It is found that the SI-Turb model has a stronger predictive capability than the SI-Wiebe model, as it can predict combustion performance at various operating conditions using a single set of parameters. Moreover, SI-Turb can effectively capture the combustion and performance changes caused by altitude. As the altitude changes, the 2DO contour maps showed consistent trends in power, peak cylinder pressure, and BSFC. The 2DO visualization method successfully highlights the tradeoffs between different objectives and reveals the impact trends of the factors. In single-point optimization for maximizing fuel economy, the results obtained through 2DO show that the BSFC can be improved by up to 9.25 % within the altitude range of 0–4000 m. At higher altitudes, achieving better fuel economy requires the use of earlier injection and ignition timings in combination with a lambda value greater than 1.05. The proposed methods can enable effective selection of factor settings based on altitude-specific design requirements.