Adopting new type diagonal regression neural network to identify the system model

the parameters of PID controller have been set

and the multi -variable decoupling control being realized. The outlet temperature and inlet negative pressure control system of ball mills in thermal power plant has been designed by using said method

and simulation study being carried out. Results of simulation show that the control system can be decoupled automatically

having high stability

strong robustness

rapid response speed to regulation

small dynamic deviation

and without static deviation.