Polycrystalline silicon wafers are the main sources of solar cells. In this study, we used numerical analysis to optimize the growth of polycrystalline silicon ingots in a directional solidification system (DSS). The numerical analysis method was based on the finite volume method (FVM), using Fluent software as a tool to analyze the thermal field of the directional solidification process through the solidified-molten model, considering the effects of heat conduction, heat convection, and heat radiation. We used a GT Solar Ltd. directional solidification crystal growth furnace as the prototype for the finite element model analysis. We used triangles and quadrilaterals as the cutting meshes to simulate the two-dimensional structure using Fluent. The software and operating processes were divided into three processes: pretreatment, numerical calculation, and postprocessing. We used the simulation technology to understand the effect of the temperature profile on the growth characteristics of polycrystalline silicon wafers during the silicon ingot growth process, allowing us to optimize the crystal growth process.