Introduction  With the development of artificial intelligence technology, combustion optimization schemes based on intelligent optimization algorithms emerge endlessly, such as GA, PSO, FPA, and so on. When developing the real-time combustion optimization system based on the intelligent power plant platform, it is necessary to weigh and choose the best technical scheme.

  Method  Aiming at the problem that the weight assignment of the traditional TOPSIS method is highly subjective, an optimal weight fusion method was proposed to improve it, and the improved TOPSIS method was used to establish an evaluation system for the intelligent combustion optimization scheme of thermal power units. The six schemes of GA, AGA, PSO, FPA, CSO, and GSA were evaluated comprehensively from three aspects: optimization effect, optimization period, and reliability.

  Result  The results show that through the double verification of the MCD index analysis system and the comparison with the traditional method, the weight of the improved TOPSIS method is consistent with the importance order of the MCD index, and compared with the traditional TOPSIS method, the pros and cons of each scheme can be distinguished better. The results are more in line with the requirements of the actual production process and have the advantages of strong objectivity and good accuracy.

  Conclusion  The research can provide a valuable reference for the selection of combustion optimization schemes for thermal power units.