Intelligent control of coke oven

Abstract

Coke oven is a complex plant with the characteristics of large time-delay, strong non-linear, multivariable coupling and changeable parameters. The longitudinal temperature was affected by many reasons, the control principle of combining the intermittent heating control with the heating gas flow adjustment was adopted. Intelligent control methods, namely fuzzy control and neural network, were proposed to establish intelligent control strategy and model of coke oven, which combined two feedback control, one feed forward control and intelligent control. Initial gas flow was given by heating supplied feed forward model according to coking mechanism, and carbonization index feedback model was proposed in the model to revise the goal temperature to control coking management of coke oven. Flue temperature soft measurement model based on linear regression and neural network was built to supply temperature feedback control. According to artificial operation and actual condition, fuzzy controller was designed. Intelligent control methods were used to adjust stopping heating time and heating gas flow. The practical running results indicate that the system can achieve heating intelligent control of coke oven, stabilize production of coke oven, effectively improve quality of coke and decrease energy consumption, and has great practical value.