This paper presents a segmented dynamic modeling approach for the tail of a beaver-like underwater robot, combining hydrodynamics and material mechanics to develop a flexible dynamic modeling theory. The study aims to evaluate the motion performance of the robot's tail through a performance index that considers propulsion efficiency and motion stability. Simulations and experiments were conducted to verify the dynamic model and assess the tail's motion performance under different swing amplitude coefficients. The optimal swing amplitude coefficient was found to be 2, which is crucial for improving the robot's propulsion ability and motion stability. The tail of the beaver-like robot is composed of a tail root and a tail body, divided into segments for dynamic modeling. The tail's motion is analyzed using a dynamic coordinate system, with equations describing the deflection and rotation of each segment. The hydrodynamic force on the tail is calculated based on the robot's dynamic interpolation formula, and the force at the tail root is determined through internal calculations. A motion performance index $ S_H $ is proposed to evaluate the tail's motion performance, combining propulsion efficiency and motion stability. The index is calculated using the propulsion and lift forces over time, with the results showing that the optimal tail motion trajectory is achieved when the amplitude coefficient is 2 and the time coefficient is 2. This trajectory aligns with the bionic tail trajectory, demonstrating superior propulsion efficiency and stability. Theoretical calculations, CFD simulations, and underwater experiments were conducted to validate the dynamic model and motion performance index. The results show good consistency between the theoretical model, simulation, and experimental data, confirming the correctness and effectiveness of the dynamic model. The study contributes to the development of dynamic modeling and motion analysis for other underwater biomimetic robots.This paper presents a segmented dynamic modeling approach for the tail of a beaver-like underwater robot, combining hydrodynamics and material mechanics to develop a flexible dynamic modeling theory. The study aims to evaluate the motion performance of the robot's tail through a performance index that considers propulsion efficiency and motion stability. Simulations and experiments were conducted to verify the dynamic model and assess the tail's motion performance under different swing amplitude coefficients. The optimal swing amplitude coefficient was found to be 2, which is crucial for improving the robot's propulsion ability and motion stability. The tail of the beaver-like robot is composed of a tail root and a tail body, divided into segments for dynamic modeling. The tail's motion is analyzed using a dynamic coordinate system, with equations describing the deflection and rotation of each segment. The hydrodynamic force on the tail is calculated based on the robot's dynamic interpolation formula, and the force at the tail root is determined through internal calculations. A motion performance index $ S_H $ is proposed to evaluate the tail's motion performance, combining propulsion efficiency and motion stability. The index is calculated using the propulsion and lift forces over time, with the results showing that the optimal tail motion trajectory is achieved when the amplitude coefficient is 2 and the time coefficient is 2. This trajectory aligns with the bionic tail trajectory, demonstrating superior propulsion efficiency and stability. Theoretical calculations, CFD simulations, and underwater experiments were conducted to validate the dynamic model and motion performance index. The results show good consistency between the theoretical model, simulation, and experimental data, confirming the correctness and effectiveness of the dynamic model. The study contributes to the development of dynamic modeling and motion analysis for other underwater biomimetic robots.