Abstract: Variable-rate fertilization is an important part of precision agriculture. Nonlinearity, large inertia and time-varying parameters are the key factors affecting the accuracy and steady-state performance of variable rate fertilization control system. PID control algorithm is widely used in the field of industry and agriculture because of its simplicity and convenience, but it is often difficult to achieve the desired control effect. Gray wolf optimization algorithm（GWO） is a swarm intelligence optimization algorithm with few parameter settings and good convergence performance, but it is easy to fall into local optimal solution in the iterative process. Based on this, this paper introduces genetic crossover and mutation operators into the standard GWO algorithm, combined with the good point set method, proposes an improved new grey wolf optimization algorithm（GGWO）. Applying the improved genetic grey wolf optimization algorithm to the PID control of the water and fertilizer integrated control system. Taking the liquid fertilizer control system as the research object, the corresponding mathematical model of negative feedback control system is established. Three different control methods, conventional PID control, GWO based PID control and GGWO based PID control is simulated with MATLAB. Finally, the system performance indexes under each control method are compared and analyzed. The simulation results show that the PID control based on GGWO is superior to the other two control methods in the performance indicators of the system, such as rise time, adjustment time and fitness value, and achieves better control effect in the accuracy, uniformity, robustness and steady-state performance of the system, which not only meets the operational requirements of precision agriculture, but also lays a foundation for subsequent research.