Pyrocumulonimbus (pyroCb) are fire-triggered or fire-augmented thunderstorms and can by transporting a large amount of smoke particles into the lower stratosphere. With satellite remote sensing measurements, the plumes from pyroCb events over British Columbia in 2017 were observed in the lower stratosphere for about 8-10 months after the smoke injections. Several previous studies used global climate models to investigate the physical parameters for the 2017 pyroCb events, but the conclusions show strong model dependency. In this study, we use Energy Exascale Earth System Model (E3SM) atmosphere model version 1 (EAMv1) and complete an ensemble of runs exploring three injection parameters: smoke amount, the ratio of black carbon to smoke, and injection height. Additionally, we consider the heterogeneous reaction of ozone and primary organic matter. According to the satellite daily observed aerosol optical depth, we find that the best ensemble member is the simulation with 0.4 Tg of smoke, 3% of which is black carbon, a 13.5 km smoke injection height, and a 10-5 probability factor of the heterogeneous reaction of ozone and primary organic matter. We use the Random Forest machine learning technique to quantify the relative importance of each parameter in accurately simulating the 2017 pyroCb events and find that the injection height is the most critical feature. Due to the long lifetime and wide transport of stratospheric aerosols, the estimated e-folding time of smoke aerosols in the stratosphere is about 188 days, and the global averaged shortwave surface cooling is -0.292 W m-2 for about 10 months.