A dynamic model for the flue gas desulfurization tower in coal-fired power plants is proposed. The model is validated with the data from a 350 MW unit. The penetration theory is adopted in modeling. The movement and the size of the slurry droplets, the temperature and the flow rate of the flue gas, as well as the chemical reactions have been taken into account. Three parameters in the model are found to have a critical impact on the model accuracy, i.e., the droplets’ initial velocity, the number of droplets leaving all nozzles in a single spray layer per unit time, and the liquid side mass transfer coefficient of sulfur dioxide. These parameters need to be identified periodically in real applications. The quantitative relationship between pH and liquid side mass transfer coefficient is given. The main hypotheses and parameters of the model were discussed in order to quantify their respective influence. The model may be used to simulate the desulfurization process across the entire load range. Specifically, it may be used to predict the desulfurization efficiency online, so that applied potentially for the operational optimization.