Active magnetic bearing (AMB) is a mechanical device that uses electromagnetic coils to convert electrical energy into magnetic energy to support the rotating shaft. In the field of contemporary industrial manufacturing, compared with traditional bearings, active magnetic bearings are often used in occasions with special requirements for speed, accuracy and operating environment due to their advantages such as no mechanical friction loss, extremely high operating speed, low power consumption and low noise. An optimized LQR control method based on genetic algorithm (GA) was adopted, utilizing the advantages of quick response and small overshoot of the optimal control to realize rapid suspension and stable position of the bearing. Meanwhile, genetic algorithm was used to optimize the Q matrix and R matrix of the optimal controller. In the genetic algorithm, the optimal solution was searched by simulating the process of biological evolution. The fitness of the genetic algorithm was set according to the deviation of the system to obtain the approximate optimal solution of Q and R matrix. At last, the simulation was realized on MATLAB platform.