A large amount of energy is wasted in integrated desulfurization and denitration device of coking flue gas during coking reverse process. In order to solve this problem in the unique coking reverse process

causes to NOx concentration change in the process were thoroughly analyzed. A modeling method for NOx concentration disturbance at inlet and outlet in the reverse process was proposed by data driven identification. Verification was performed from actual data measurement on two 55-hole and 6-meter top charging coke ovens and the relationship model of NOx concentration loss at inlet and outlet was obtained. A feedforward control system was designed for feedforward control rates of O3 under different NOx concentration settings. Simulation results provided cost savings of power consumption

which indicates necessity of the work.