The feed coal quantity of boiler is calculated according to the design coal.However

due to the complexity of feed coal type and the large fluctuations of coal calorific value in power plants

accurately controlling the amount of coal into the furnace has become a difficult problem.Thus

this paper designed the self-adaptive coal quality feedforward correction

and established a prediction model for feed coal quantity

by using the coefficient of power to coal as a starting point

on the basis of the combined prediction model and variance reciprocal method.The optimal weight coefficient of the power to coal coefficient in the last one hour to the current power to coal coefficient was calculated

so the coefficient of power to coal and the amount of feed coal into the furnace can be predicted.Moreover

taking No.4boiler of the 660 MW unit in Guohua Cangdong Power Plant as the example

the feed coal quantity was predicted using the above model.The results show that

using the real-time predicted value as the actual feed coal amount

the average accuracy can reach up to 97%

indicating the proposed prediction model has a good prospect.