As space technology advances, the exploration of small celestial bodies has become a major research focus, necessitating more stringent requirements for microgravity simulation technology. The microgravity simulation device featuring a magnetically induced zero-stiffness structure is utilized in ground simulation experiments for small celestial body detectors. This device integrates a magnetoelectric coupling system, motor control system, and stiffness coupling system to achieve a quasi-zero stiffness state under high load conditions. However, its complex structure and high nonlinearity complicate the tuning of control parameters. To tackle this issue, this paper presents a tuning method that merges model identification and parameter optimization for the magnetically induced zero-stiffness microgravity simulation device. The proposed approach employs a combination of pseudo-random signals and the instrumental variable method for model identification of the microgravity system. Subsequently, genetic algorithms are used to optimize the control parameters, thereby shortening the tuning process. Simulation and experimental results indicate that the control accuracy of the microgravity simulation system surpasses 98%, demonstrating the method’s feasibility and effectiveness.