Aiming at the problems of large crack, poor quality and low molding rate of straw particles produced by straw compactor due to its low moisture content in the field, a moisture content control model was constructed, and a moisture content control system of straw compactor was proposed based on SSA-Smith-LADRC. The system used smith predictor to solve the time-delay problem of the system, and used sparrow search algorithm (SSA) to optimize the parameters of linear active disturbance rejection controller (LADRC) to achieve accurate control of moisture content. The simulation experiment showed that SSA-Smith-LADRC had no overshoot, and the adjustment time was 1.53s. After the water content control system was stable, the time to recover from interference was 0.62s, and the overshoot was 0. After the stable state was restored again, there was no oscillation. The experiment in the factory showed that the system had no overshoot, the maximum error was 2.0%, and the average error was 1.5%. The experiment in the filed showed that when the moisture content of straw was not regulated in normal operation, the shaping rate of straw particles was 62.4%. When the moisture content of straw was controlled by manual experience, the error of moisture content was changed greatly, the average error was 9.83%, and the molding rate was 82.1%. When the water content control system was regulated, the average error of water content was 2.82%, and the molding rate was 93.4%. Compared with Smith-PID, SSA-Smith-LADRC reduced the adjustment time by 4.05s, overshoot by 86.5%, maximum error by 42.0%, minimum error by 50.8%, and mean square error by 60.2%, respectively. The moisture content control system proposed can effectively track the change of straw moisture content, improve the molding rate of straw particles.