Abstract: Refractory high-entropy alloys (RHEAs) have attracted considerable attention due to their outstanding mechanical properties. However, the influence of their microstructural behavior on macroscopic mechanical performance remains poorly understood. With the increase of study on material micromechanical behaviors, the crystal plasticity finite element methods become essential tools for uncovering the underlying mechanisms of crystalline materials. Since crystal plasticity constitutive models involve numerous complex parameters, a thorough analysis of these parameters is critical for a deeper understanding of the micromechanical behaviors of alloys. The crystal plasticity model used in this study incorporates the Peierls stress, which accounts for the short-range potential barriers of the material, thereby enabling a more accurate simulation of its strain-rate behavior. Through experimental design and range analysis, the key constitutive parameters affecting the alloy’s mechanical properties were identified. Univariate analysis was then employed to clarify the specific effects of these critical parameters on the mechanical characteristics of the material. For parameter inversion, an optimization-based approach was developed, combining the support vector regression with optimization algorithms. This method effectively inverts crystal plasticity constitutive parameters from macroscopic mechanical testing data. For the cast TiZrNbV alloy, a set of optimal parameters was successfully inverted, and the agreement between simulation results and experimental data validated the method’s effectiveness. This study provides valuable insights for predicting the mechanical behaviors, guiding material design, and optimizing the performances of refractory high-entropy alloys.