Multivariable hybrid intelligent optimal control for coke oven heating process

Coke oven heating process in coking production is a synthetic process, whose product key index, i.e. flue temperature, is hard to measure online, and the optimal control of such process is very difficult. Based on analysis of its dynamic characteristics and manual operation mode, a multivariable intelligent optimal control method is developed, combining intelligent method, feedforward control and feedback control. The target is the optimization of integrated production indices including coke quality and amount of consumed energy per ton coke. The proposed method is composed by a control loop setting model of flue temperature, hybrid prediction model of the techniques indices, a fault work-condition diagnosis model, feedforward and feedback compensators. The settlement value of control loops of the process are adjusted online automatically according to the targeted integrated production indices and variations of boundary conditions, so the optimal control is achieved. The practical running indicates that the system can realize intelligent control of coke oven heating, stabilize production of coke oven, effectively improve quality of coke and decrease energy consumption, and has great practical value. The application results have proven the effectiveness of the proposed method.