Modeling of outlet temperature of the first-stage cyclone preheater in cement firing system using data-driven ARMAX models

摘要

A model of outlet temperature of the first-stage cyclone preheater in the cement firing system based on the data-driven autoregressive moving average (ARMAX) model is proposed to resolve the problem of temperature control based on mathematic model. First, the Butterworth low pass filter and the normalized processing are used to process the cement firing system data. Then, the modeling input variables are selected by Pearson correlation analysis. The order determination method based on the final prediction error is used. Finally, the parameters of ARMAX model are identified by recursive maximum likelihood algorithm and the obtained model is verified using residual analysis method. In the simulation experiment, the model input variables are more reasonable and the established ARMAX temperature model is compared with principal component regression and partial least square regression model, which verifies that the ARMAX temperature model can accurately and stably estimate the outlet temperature of the first-stage cyclone preheater according to the real-time data, thus improving the prediction accuracy.