Research Of Temperature Multi-objective Optimation And Control Method In Coke Oven Flue

The coking-blending is a complex industrial process. The control of the Coke oven tempreture plays an important part in Coke production. If the tempreture is too high, the energy consumption of unit will increase. It is very easy to appear a phenomenon of time-consuming, laborious. If the tempreture is too low, coke dose not reach maturity within the coking time. It will affect the qulity of the coke. So, how to set a reasonable coke oven tempreture, and to make sure the Coke oven tempreture stable in the set point is an urgent problem of realizing optimization control of the coking process.Now, seting the value of the temperature depends on experience. It is lack of theoretical guidance and is hard to adjust the change of actual operating mode. So, we eastablish optimal model of temperature, including the models of yield, quality and energy compution which are connected with the temperature. We can get the set value of the temperature according to the optimal model. First, we use the gray association analysis method to analyze the correlation of coking production, quality, heat comsumption and their influencing factors to determine the input of the models. Second, we use RBFNN to build quality, production and energy comsumption model. Last, with the maximum coke yield and minimum energy consumption as the optimization objective, with coke quality and technological requirements as the constraint conditions and with the target flue temperature as the decision variables, calculate a reasonable temperature.After getting the set point, it needs to designe an effective controller to make the temperature stabilizing in the set value. The coke oven temperature control system is a strong nonlinear, large time delay and intense interference system. PID control dose not meet the demand of the control. According to the characteristic of the controlled object, we use an improved implicit GPC, introducing the soft coefficient matrix and making up for the control increment to decrease the calculation. In the meanwhile, it can restrain overshoot of the output. Finally, through simulation, we compared the effect of PID control and the improved implict control. It conforms that the improved implict GPC is better than PID control algorithm.