The artificial intelligence (AI) computing platform that is close to human intelligence could not be set up by general industrial computers, or even general-scale computer clusters. On this basis, this paper puts forward a basic framework of smart power generation technology for thermal power units. A ‘smart power generation technology center’ is built in the technology institute of a certain scale power generation group, which concentrates on the research and development of intelligent technology; Real-time data and information are transmitted to the center by various power plants and generating units, the latter can obtain technological support on smart control, smart operation and maintenance, and smart management from the center. Then the on-line monitoring technology of three-dimensional temperature field in the furnace, the monitoring technology of combustion load in the furnace and the control technology of quickly overcoming the disturbance of fuel calorific value, the air-coal-water independent decoupling, precise feed-forward control and optimization technology of boiler operation based on operation data analysis, which were developed by the author's team, are introduced briefly. These technologies have been successfully put into operation in a 300 MW and a 600 MW coal-fired power generation units, and have achieved the goal of reducing coal consumption and nitrogen oxide emission in the furnaces at the same time, laying a good foundation for the research and development of smart power generation technology.

Keywords： artificial intelligence ; smart power generation ; combustion optimization ; combustion monitoring ; big data analysis ; nitrogen oxide emissions