Abstract: As one of the core safety guarantee devices of nuclear reactors, the control rod hydraulic drive system (CRHDS) is crucial for ensuring the precision of reactor power regulation and safety performance under accident conditions, as its operational stability and response reliability are directly associated with the aforementioned key indicators. The performance of CRHDS directly affects the operational safety and economy of nuclear reactors. Therefore, in-depth research on its working mechanism and performance characteristics holds significant engineering practical value. The working principle of CRHDS was first analyzed in this paper. To accurately grasp the actual operational performance of the system, a full-scale cold performance test of the drive system was conducted, covering the lifting, insertion, and rapid rod-drop transient processes of the control rod assembly. During the test, the nuclear-grade equipment test specifications were strictly followed. High-precision pressure sensors, electromagnetic flowmeters, and displacement sensors were installed at key nodes of the system’s hydraulic circuit as well as the inlet and outlet of the drive mechanism, enabling millisecond-level synchronous and real-time acquisition of core physical parameters such as system pressure, flow rate, and control rod displacement, which ensures the integrity and accuracy of the test data. Based on the collected massive valid test data, an in-depth analysis of the variation laws and intrinsic action mechanisms of the relevant physical parameters of the CRHDS was conducted in this paper. The research results indicate that during the lifting and insertion processes of the CRHDS, the variation trends of system pressure and flow rate are completely opposite. Specifically, the fluctuation amplitude of parameters during the insertion process is more moderate than that during the lifting process. In addition, the change in throttle port opening can be derived from the variation law of system flow rate. For instance, during the lifting process, the throttle port opening first increases rapidly, remains constant during the control rod movement phase, and finally decreases to the initial state when the movement stops; the variation trend of the throttle port opening during the insertion process is opposite to that during the lifting process. The findings of this test study not only verify the rationality and feasibility of the CRHDS design scheme but also provide detailed and reliable performance data support. They lay a solid foundation for the structural optimization, key parameter matching, control strategy improvement, and fault diagnosis model construction of the CRHDS, and have important guiding significance for enhancing the overall operational safety and economy of nuclear reactors.