Abstract: Abstract: This study investigates the coordinated frequency regulation control of large thermal power units considering inertia. Addressing the challenges of reduced system inertia and heightened frequency stability risks under high renewable energy integration, while existing primary frequency regulation systems struggle to fully utilize their inertial support potential, we conduct research on optimized control strategies. By establishing dynamic response models, we reveal the coupling relationship between inertial response and primary frequency regulation across different time scales. The proposed multi-scale signal decomposition and differentiated control strategy decomposes frequency deviation signals into high, medium, and low-frequency components. Through implementing feedforward control based on frequency rate of change, adaptive droop control, and signal filtering techniques, the study achieves organic synergy between inertial support and active frequency regulation during dynamic processes, enhancing the unit''s comprehensive suppression capability for full-frequency band power disturbances. Case studies on large thermal power units demonstrate that the proposed method effectively suppresses maximum frequency deviations and rate of change, accelerates frequency recovery, and significantly improves grid frequency dynamic quality, providing an effective solution for enhancing frequency stability in next-generation power systems.