| Cement is the supporting industry of the construction industry. It is the indispensable adhesive in concrete. New dry process cement with the core of precalcining technique is the currently most advanced cement production technology in the world. Cement clinker free lime (f-CaO) content which is produced by calcining process of cement rotary kiln, would affect the quality of cement. Poor quality of cement will not only have an impact on the building materials, but also a waste of resources. Many years of industrial practice is concluded that f-CaO content should be controlled in a reasonable range (0.5-2.0%), to ensure the cement quality. However, now, the monitor of f-CaO content is still in the stage of artificial test, and there is no specific instrument to detect on line. That could lead to big lag and be not conducive to quality control. So, how to acquire the real-time f-CaO content, and laying the theoretical foundation for on-line monitor and control of production quality is the research focus of this thesis.This thesis’s topic was selected from Yunnan province institute special item, "Research and development and application demonstration of f-CaO on-line soft-measuring and temperature control system in cement production process". For new dry cement production line with 2000t/d in a Yunnan cement factory, with the help of soft-measuring technology to realize on-line prediction of f-CaO content, solving the problem which f-CaO content is difficult to measure online at present.Firstly, this thesis on the basis of deeply understanding the cement process, analyzing the influence factors of f-CaO content, determining the f-CaO soft-measuring model structure, by using three kinds of soft-measuring methods: multiple linear regression (MLR), multiple stepwise regression (MSR) and recursive partial least squares (RPLS), respectively established cement clinker f-CaO soft-measuring model and carried out the model validation. Secondly, in order to improve the model’s predictive accuracy, the prediction data of model was carried out correction method studied, getting a better correction result, and by comparing the prediction effect of three kinds of model, MLR, MSR, RPLS, determined the model algorithm which could better adapt to the cement production is RPLS algorithm. Meanwhile, in order to helpfully realize the optimal closed-loop control of the outlet temperature on cement decomposing furnace in the future, this thesis according to the relationship between the f-CaO soft prediction value and the outlet temperature of decomposing furnace, further developing the research on optimization calculation of the best set-point value for decomposing furnace outlet temperature, obtained the satisfactory result. Finally, to achieve that making f-CaO soft prediction algorithm apply to engineering, On the basis of GE-Cimplicity software, developing the f-CaO soft prediction monitor screen, and by OPC interface and VBS script, successfully realized the data exchange between f-CaO soft prediction algorithm based on Matlab and monitor screen on based on Cimplicity.The research of this thesis lays the certain theoretical foundation for f-CaO content’s on-line measurement and closed-loop control of cement clinker quality, and providing an effective way for the project implementation research. That would have important practical significance for the cement industry.
|
PDF Full Text Request