The calculation of unit energy consumption in the actual production process of slabs is in-fluenced by factors such as raw material and fuel conditions,operating processes,and steel grades,and the mapping relationship between each factor and the unit energy consumption of slabs is relatively complex.In the paper,the prediction model of slab unit erergy consumption is established by using BP neural network,and the actual production data of slab heating furnace is taken as the research object,and 17 influencing factors involved in the heating process are used as input variables to establish a slab unit energy consumption calculation and prediction model.Combining trial and error methods,a rea-sonable BP neural network structure is determined as follows:the number of input layer nodes is 17,the number of hidden layer nodes is 10,and the number of output layer nodes is 1.The prediction re-sults show that the predicted unit energy consumption trend is consistent with the actual value,and the predicted root mean square error was only 0.181 GJ/t,The overall accuracy of the model can reach 92.06%.
heating furnaceBP neural networkslabtotal influencing factorunit energy con-sumptionprediction model
NETL
NSTL
万方数据
