Abstract
The vast majority of combustion diagnostics methods are based on indirect inference based on measuring process parameters. At the same time, control and diagnostics cover ever smaller parts of the system. This requires the use of new measurement and diagnostic methods. The article describes flame diagnostics based on optical measurement methods and processing using artificial intelligence. The integration of optical measurement methods and artificial intelligence in flame diagnostics opens up new opportunities for improving the efficiency and safety of combustion processes. This area requires further research and development, but already demonstrates significant potential for application in various industries