Pressure Control System of Cruise Engine Room Based on BP Neural Network PID

In order to achieve accurate and stable control of cruise engine room air pressure, Comsol builds a room simulation model of cruise engine room, and then systematically identifies the mathematical model between fan speed and air pressure. Finally, aiming at the nonlinear, parameter time variability and time delay problems of cruise air pressure control system, The direction propagation neural network combined with conventional proportional integral and differential is proposed to control the temperature and pressure of the cruise engine room. BP neural network PID controller is designed, and the parameters of conventional PID controller are automatically adjusted online by using the online self-learning ability of BP neural network. Simulink model is built for simulation and verification. Compared with conventional PID control and fuzzy PID control, the response time of BP neural network PID control is shortened by 76.47% and the overshoot is reduced by 4.8% compared with fuzzy PID control. Compared with conventional PID control and fuzzy PID control, BP neural network PID adjustment time and overshoot are reduced, its dynamic response ability is improved, and it shows better adaptive ability.