Aiming at the weaknesses of conventional PID control such as hysteresis and low accuracy in nonlinear and large inertial systems, a control system combining neural network and prediction algorithm is constructed. The pre-control algorithm is adopted, which makes full use of the rolling optimization and feedback correction characteristics of predictive control, and uses the neural network to establish the dynamic model of the system as the prediction model of the prediction controller, which realizes the adaptive control of the large lag system, which has real-time control and prediction performance, effectively improves the control accuracy and reliability, and enhances the stability. The field operation results show that the method used in the air purifier system has good effect and is easy to promote.