For the randomness and volatility of wind energy, wind power generation systems are prone to power fluctuations, power fluctuations can be stabilized by using superconducting magnetic energy storage (SMES) and battery (BESS) hybrid energy storage methods. An improved hybrid genetic algorithm for variable parameter state-of-charge control strategy is proposed. The algorithm is improved by the idea of adaptive learning, and the convergence speed and accuracy of the algorithm are also improved. The state of charge residual and state partition of state of charge are used as the objective function and boundary conditions of the improved hybrid genetic algorithm. The obtained target result as a correction value to correct filter time constant to achieve power secondary distribution. The simulation model is built in MATLAB/Simulink to verify the effectiveness of the control strategy. The proposed control strategy can optimally match the SMES and BESS output at any time, at the same time, it can reduce the battery charging and discharging depth and improve the smoothing effect on wind power fluctuation, then it can also effectively improve the service life of the hybrid energy storage system.