Prediction of Temperature in Firing Zone of Cement Rotary Kiln with Nonlinear Autoregressive Neural Network Model

Abstract

The core equipment in cement production is cement rotary kiln, and the quality of cement clinker is mainly determined by the temperature of rotary kiln. Studying the modeling method of firing zone temperature enables operators to know the prediction results in advance, which is meaningful for stabilizing working conditions, improving quality and reducing energy consumption. Aiming at this problem, this paper proposes a non-linear autoregressive neural network modeling method for firing zone temperature, and conducts experimental research. This paper firstly introduces the cement production process and process parameters, then determines the input parameters of the model through data preprocessing and gray correlation analysis, and finally establishes a nonlinear autoregressive neural network model, and the sintering zone temperature of rotary kiln was predicted and verified in a single step. The experimental results indicate that the prediction of the temperature of the firing zone of the rotary kiln through the nonlinear autoregressive neural network has a good prediction effect and can play a guiding role in the production on site.