A large amount of pollutants is emitted into the air in coal-fired power plants. Among them, SO2 can cause various diseases to the human respiratory system. At the same time, it will react with rain to form acid rain during the precipitation process, which will cause serious harm to the ecological environment of our country. It is very important to control SO2 emission. Flue Gas Desulfurization (FGD) technology is one of the important means to control SO2 emissions in industry. The characteristics, principles, process flow and influencing factors of desulfurization efficiency of the two commonly used FGD technologies in coal-fired power plants: The circulating fluidized bed furnace desulfurization technology and the limestone-gypsum wet FGD technology were introduced. And the research status of optimization methods of FGD technology in coal-fired power plants was summarized. The methods of technology optimization include process modeling optimization, automatic control system optimization and system operating conditions optimization. In view of the complexity of FGD working conditions and the lack of overall optimization research, it is proposed that artificial intelligence technology can be applied to the optimization of the complexity of flue gas desulfurization technology. In addition, the traditional technical evaluation method focuses on the evaluation of technical and economic performance,and the impact of a large number of greenhouse gases released by FGD technology on the environment can′t be comprehensively analyzed. Under the background of national energy conservation and emission reduction, in view of the limitations of traditional technology evaluation methods, it is proposed that the method of Life Cycle Assessment (LCA) can be adopted for FGD system to realize the measurement control of carbon emissions and other harmful substances emissions during its entire life cycle, and conduct a comprehensive environmental impact assessment. The application of Life Cycle Assessment in flue gas desulfurization of coal-fired power plants is summarized. LCA combined with process simulation is put forward to quickly obtain the evaluation results of different processes and improve ability to select, optimize and make decisions on different technologies.