Ethylene plays a crucial role as an intermediate component in the cracking and combustion processes of large molecular alkane and olefins. In this article, the laminar flame speed of ethylene–air mixtures was measured using the heat flux method. The mechanism of ethylene was simplified by utilizing the error propagation directed relationship graph (DRGEP) and sensitivity analysis (SA), and the Arrhenius pre‐exponential factors for 20 selected reactions in the skeletal mechanism were optimized using the particle swarm optimization (PSO) algorithm. Finally, an ethylene optimization mechanism including 39 species and 85 reactions was obtained. The prediction results for flame speed, ignition delay time, and species concentration were compared with experimental data and other mechanisms, covering a wide range of temperatures (298–1725 K), pressures (1–22.8 atm), and equivalence ratios (0.5–2.0). The findings demonstrate that the optimization mechanism not only improves the prediction results of laminar flame speed in the rich combustion zone and low oxygen environment but also enhances the prediction accuracy of the ignition delay time at high pressure and in the lean combustion zone, as well as the prediction accuracy of C2H4 and H2O radicals. In conclusion, the optimized mechanism exhibits higher accuracy and broader applicability.