The performance of a gas turbine directly affects its energy conversion efficiency and operation stability rate, while the optimization of the internal flow field is the key to improving its performance. Computational Fluid Dynamics (CFD), as a powerful numerical simulation tool, plays an important role in optimizing the gas turbine flow field. In this paper, the turbine of the gas turbine is taken as an object, and CFD technology is used to analyze the distribution of the flow field inside the turbine and to realize the optimization design of the gas turbine in this way. The finite volume method (FCM) based discretization of the N-S governing equations is selected, and the RNG k-ε turbulence model is used for numerical simulation. The mounting angle, geometric angle, and mid-arc are used to optimize the aerodynamic joint of the turbine’s dynamic and static blades based on the results of the 3D flow field analysis. The error of the RNG k-ε model in the efficiency characteristic line is smaller than that of the SST model and the EARSM model, and its maximum relative errors with the experimental values of efficiency and flow rate are 3.4% and 4.6%, respectively, which are within the reasonable range, indicating that the reliability of the numerical calculation method in this paper meets the requirements. Compared with the initial design of the turbine, the mass flow rate and total efficiency of the optimized turbine in this paper have been improved by 0.88% and 4.93%, respectively, and the expansion ratio has been reduced by 6.15%, which is a better optimization effect, and lays a practical foundation for further optimization of the design of the gas turbine.