In order to reduce energy consumption and improve the stability of cement burning system production, it is necessary to
conduct in-depth analysis of the cement burning system, control the operation state and law of the system. In view of the
rotary kiln consumes most of the fuel, we establish the simulation model of the cement kiln used to find effective control
methods. It is difficult to construct mathematical model for the rotary cement kiln as the complex parameters, so we
expressed directly using neural network method to establish the simulation model for the kiln. Choosing reasonable state
and control variables and collecting actual operation data to train neural network weights. We first in-depth analyze
mechanism and working parameters correlation to determine factors of the yield and quality as the model input variables;
then constructed cement kiln model based on BP and Elman network, both achieved good fitting results. Elman network
model has a faster convergence speed, high precision and good generalization ability. So the Elman network based model
can be used as simulation model of the cement rotary kiln for exploring new control method.
|