With the rapid development of industry and improvement of environmental protection requirements, the control of nitrogen oxide (NO_x) emissions has become an important issue in air pollution control. Although the traditional denitrification processes such as SCR have been relatively mature, most of them still use rough manual control or automatic and semi-automatic control methods with a low refinement, which have many defects including large emission fluctuations, large resource consumption and high labor intensity. In recent years, with the rapid development of artificial intelligence, machine learning and big data analysis technologies, intelligent technologies have been gradually introduced into the control of industrial gas SCR denitrification process to overcome the shortcomings of traditional control methods, and have achieved certain results. The research progress of intelligent control models for gas denitrification process in recent years is reviewed, the optimal control methods for gas denitrification based on such advanced technologies as big data, artificial intelligence, and machine learning are introduced; NO_x concentration prediction and denitrification control optimization are carried out through BP neural network, deep learning (such as LSTM, CNN) and other methods; the research on optimization method of BP neural networks (such as combining gray wolf algorithm, particle swarm algorithm, etc.) is reviewed; the advantages and control effects of PID-based fuzzy control and adaptive control models such as two-degree-of-freedom strategy is compared; the application status and challenges of intelligent control technology for denitrification process in industry are discussed; finally, the future development trend and research direction of intelligent control technology for denitrification process are prospected, in order to provide reference for industrial gas pollution control.