A machine learning based intelligent control method for wide-load optimization of CO₂capture process in coal-fired power plant

Under the background of deep peak-shaving in coalfired power plants, the flexible control of the CO₂ capture process is of great significance for achieving efficiency decarbonization under wide-load range conditions. In this paper, an intelligent control method based on machine learning is proposed to maintain the flue gas capture rate requirements while keeping the energy consumption at a low level across a wide-load range. First, a thermodynamic model for the typical CO₂ capture process with 30 wt.% monoethanolamine (MEA) solution is established. Second, considering the key operating parameters of the CO₂ capture process, a machine learning model is trained to achieve accurate predictions of the capture rate and specific regeneration energy. Finally, an intelligent control method is proposed to minimize energy consumption while maintaining the required CO₂ capture rate across wide-load range. The results show that the energy consumption of CO₂ capture can be maintained near the minimum range using the intelligent control method, while maintaining a CO₂ capture rate with a relative error of less than 5%, achieving the flexibility, high efficiency and low carbon of the coal-fired power plant.