Abstract: To address significant overshoot, slow response, and hysteresis in traditional PID control for hydraulic support cylinders caused by strong nonlinearity, parameter dispersion, and coupled multi-source disturbances, a co-optimization control framework integrating genetic algorithms and fuzzy PID is proposed. This approach establishes 3D fuzzy rules and adopts genetic-algorithm-based multi-objective parameter optimization to construct a self-tuning nonlinear compensation mechanism. A high-fidelity electro-hydraulic coupling model is built on the AMESim-Simulink co-simulation platform, and multi-condition validations are carried out. Experimental results demonstrate that this method substantially reduces the overshoot of step response, effectively suppresses displacement fluctuations under disturbances, and significantly improves the posture control accuracy and disturbance-rejection robustness of hydraulic supports. It provides theoretical and technical foundations for the development of intelligent deep-mining equipment.