In order to improve the dust removal efficiency of the spray dust removal system, a variable particle size spray dust removal control system based on the improved BP neural network PID is designed. The system takes spray air pressure as the control variable, establishes the relationship function between spray pressure and spray particle size, optimizes the parameters of the spray pressure PID controller of the variable particle size spray dust removal system by using the BP neural network algorithm, and uses Simulink to simulate the variable particle size spray dust removal system using the traditional PID controller and the improved BP neural network PID controller respectively. Experimental results show that the improved BP neural network PID algorithm improves the response speed by 66.63% and the stability by 20% compared with the traditional PID algorithm, which greatly improves the spray pressure control effect of the variable particle size spray dust removal system.