We conduct the first microlensing simulation in the context of planet formation model. The planet population is taken from the Ida & Lin core accretion model for $0.3M_odot$ stars. With $6690$ microlensing events, we find for a simplified Korea Microlensing Telescopes Network (KMTNet) the fraction of planetary events is $2.9%$ , out of which $5.5%$ show multiple-planet signatures. The number of super-Earths, super-Neptunes and super-Jupiters detected are expected to be almost equal. Our simulation shows that high-magnification events and massive planets are favored by planet detections, which is consistent with previous expectation. However, we notice that extremely high-magnification events are less sensitive to planets, which is possibly because the 10 min sampling of KMTNet is not intensive enough to capture the subtle anomalies that occur near the peak. This suggests that while KMTNet observations can be systematically analyzed without reference to any follow-up data, follow-up observations will be essential in extracting the full science potential of very high-magnification events. The uniformly high-cadence observations expected for KMTNet also result in $sim 55%$ of all detected planets being non-caustic-crossing, and more low-mass planets even down to Mars-mass being detected via planetary caustics. We also find that the distributions of orbital inclinations and planet mass ratios in multiple-planet events agree with the intrinsic distributions.
Download