Monocrystalline silicon is the fundamental material for chip manufacturing, and its quality depends not only on the control of impurity concentration but also on minimizing crystal defects. The density of native point defects in the crystal is one of the critical indicators for assessing the quality of the crystal, which requires optimization of the thermal field, adjustment of the temperature distribution during the crystal growth process, and precise control of the V/ G ratio ( the ratio of the pulling speed to the axial temperature gradient within the crystal). This study employs the finite volume method in ANSYS Fluent software to analyze the effect of different heat shield structures on the temperature distribution during the growth of 300 mm semiconductor-grade monocrystalline silicon by the Czochralski (Cz) process. Specifically, we investigated a two-piece heat shield design, altering the structure at different angles, and simulated the temperature distribution, axial temperature gradient at the solid-liquid interface, and V/ G ratio at various stages of the pulling process (the initial stage at 400 mm, mid-stages at 800 and 1 400 mm, and the final stage at 2 000 mm). By analyzing the changes in V/ G values across these stages, a heat shield structure with V/ G values closer to the critical value ζ and better radial uniformity was found under relatively large temperature gradients, providing better conditions for reducing defect density. Additionally, by discussing the thermal history of the crystal rod, optimizing the heat shield structure to shorten the cooling cycle and providing better conditions for controlling the size of defects. The simulation results indicate that a heat shield structure with an included angle of 110°, a bottom thickness of 70 mm, and a gap of 30 mm between the inner wall of the heat shield and the crystal rod provides suitable temperature distribution conditions for the production of low-defect monocrystalline silicon.