The Study Of Improve Generalized Predictive Control In Ball Mill Application

Ball mill pulverizing systems are important equipment in the thermal powerplant, they’re widely used in domestic and international power plant. At the present,Coal grinding machine is divided into two kinds, one kind is single inlet and outlet,the other is double inlet and outlet ball mill. No matter which kind of ball mill, theyall have large time delay and strong coupling characteristics, and have many otherinterference factors, so it’s difficult for us to establish a precise mathematical model.The ball mill is a typical three into and three out of the system, for the nonlinearsystem such higher strong coupling system, how to solve the problem of modelingand automatic control is still a research focus.We analyze the working principle of ball mill system, based on the mechanismmethod, and establish the mathematical model of ball mill, add a certain disturbanceto the input variables in the mechanism mathematical model. Then we give the ballmill a step disturbance test, and establish the function model of ball mill system.When changing operating conditions of the ball mill pulverizing system islarger,the controlled object model and the model of the actual object will have largerdeviations problem, use PID will not get a good control effect. In this paper, wedesign a generalized predictive control regulator with proportional integraldifferential (PID-GPC) control algorithm to control the double input and doubleoutput ball mill pulverizing system. The newly added PID regulator is used to setvalues and actual values of deviations in real time to change, not only to enhance thetracking performance of the system, but also improve anti-interference ability androbustness.At last, through the simulation of the negative pressure operation respectively for300MW and600MW unit mill coal pulverizing system of large thermal power showsthat the algorithm has better robustness and suitable than general feed-forward decoupling PID control.