Intelligent setting control of raw meal calcination proces

In raw meal calcination process, the target value of decomposition ratio of raw meal (RMDR) is different in easy calcination stage and difficult calcination stage because boundary conditions of raw meal change frequently, where RMDR cannot be guaranteed within its desirable ranges. To solve this problem, an intelligent setting control method is proposed. This method for raw meal calcination process consists of five modules, namely a RMDR target value setting model using subtraction clustering method (SCM) and adaptive-network-based fuzzy inference system containing categorical input (C-ANFIS), a control loop pre-setting model, a feedback compensation model based on fuzzy rules, a feedforward compensation model based on fuzzy rules, and a soft measurement model for RMDR. The proposed method is realized by on-line adjusting the setpoints of control loops with the change of raw meal boundary conditions. This method has been successfully applied to the raw meal calcination process of Jiuganghongda Cement Plant in China and its efficiency has been validated by the practical application results.