Experimental Investigation and Multi-Gene Genetic Programming Simulation of Portland Clinker Burnability

Abstract

In this study, the effect of chemical composition of the raw material on the clinker burnability was studied by determination of free CaO (wt %) content of clinker. The burnability of two types of Portland clinker was investigated for silica modules of 2.3, 2.5 and 2.7 and lime saturation factor of 0.88–0.98. In addition, using the Multi-gene genetic programming (MGGP) model, the burnability of clinker was predicted. The results of MGGP model indicated that the performance of the model for predicting the amount of free CaO (wt %) was acceptable. Moreover, using MGGP, a promising correlation was introduced for accurately calculating the amount of free CaO (wt %). The performance of this correlation was compared with FL-Smidth, and it was established that the average errors of MGGP correlation and FL-Smidth equation were 2.95 and 7.45 %, respectively.