“Researchers have proposed a data-driven NOx prediction model for deep peak shaving operation of circulating fluidized bed units. The whale optimization algorithm is used to optimize the parameters of long short-term memory neural networks. The model can accurately reflect the trend of NOx emissions under large-scale variable load operation of the unit, and has a good guiding role in the operation of on-site denitrification automatic control systems, providing a solution to the stability and economic problems of denitrification control systems.”