The invention relates to the technical field of industrial
flue gas purification, and discloses a desulfurization control method based on
blast furnace gas
sulfur concentration prediction, which comprises the following steps: step 1, acquiring real-time
sulfur concentration, flow velocity, temperature, pressure and current desulfurizer feeding rate of
blast furnace gas, denoising and normalizing acquired data, and storing the data in a
database; step 2, based on the
blast furnace gas
sulfur concentration
convection-
diffusion model and the
reaction rate equation, constructing a physical constraint prediction model in combination with a long-
short term memory network, and outputting future predicted sulfur concentration; and 3, predicting a result based on the sulfur concentration. According to the method, the physical constraint prediction model based on the long and
short term memory network is adopted, the technical effect of accurately predicting the sulfur concentration of the
blast furnace gas is achieved, and compared with a prediction method purely based on experience or a traditional
time sequence model in the prior art, the defects of low prediction precision and poor generalization ability under complex working conditions are overcome.