One at a Time

Numerical simulation of the autoignition process in RCM considers volume as a function of time and incorporates the influence of heat loss. The chemistry is considered throughout the duration of simulation, including the compression stroke and the post-compression event. Experimental values corresponding to the initial pressure, initial temperature, mixture composition, and the time varying reactor geometry are specified as inputs to the simulation. The simulations intend to reproduce the experimentally observed pressure profile prior to ignition, using empirically fitted parameters. In order to understand the parameters that control the model prediction with respect to the autoignition response at low-to-intermediate temperatures, sensitivity analysis is a very powerful tool.

Global Sensitivity

To further provide insights into the potential coupling of the parameters of key reactions a global sensitivity analysis methodis is used to study the ignition delay response due to the simultaneous variation in the input factors (A factors) for the baseline model. The first step in conducting this analysis was to select the uncertain parameter. The next step was to assign an uncertainty factor to each of these input factors. A search curve of the form $$k_i=G_i [sin(\omega_i s)]$$ needs to be defined. A three-dimensional scatter plot with the axes corresponding to the A factors for three reactions are shown in Fig. below. The size of the superimposed bubbles in Fig. scales with the magnitude of the ignition delay times. In addition it seems to visually confirm that the three-dimensional sub-space corresponding to these reactions has been adequately sampled by the search curve.