Application Study Of Advanced Control Technology In Cement Calcining System

Recliner and rotary kiln are important devices in new dry process cement calcining system.The raw meal of cement is decomposed by recliner,and then enters into the rotary kiln where raw meal will be calcined.The decomposition rate of raw meal and free calcium oxide(f-CaO)in clinker,mainly determined by the temperature of recliner and the temperature of the burning zone of rotary kiln,are the quality indicators of cement calcining system.So it’s important for high efficiency production of cement to control the indexes within the range of process requirements.However,the correlation between the recliner and rotary kiln is very serious.And the calcining temperature have a strong non-linearity with the coal injection volume and feeding volume.It is difficult to measure the key variables directly,so that conventional control strategy difficultly meets the requirements of field control.Advanced control technology can solve such complex control problems effectively.In a cement 5000t/d production line,the application of advanced control technology in cement burning section is studied in this paper.The controlled target of decomposing furnace production is the decomposition rate of raw material,usually controlled by controlling the temperature of decomposing furnace.Therefore,according to the decomposition rate,the temperature setting value of decomposing furnace,is given.However,the decomposition rate of raw meal can not be directly measured at the scene.A soft sensor model by T-S fuzzy neural network based on subtraction clustering and FCM is established to predict raw material decomposition rate in this paper.After determining the the setting value of decomposition temperature,a supervisory controller based on RBF neural network is adopted to adjust the feed volume of kiln tail.MATLAB simulation shows that the controller has small overshoot and fast response speed.In the process of rotary kiln production,the temperature of firing zone determines the content of f-CaO in clinker,and its variations can also reflect the production condition of rotary kiln.In order to solve the problem that the temperature of the firing zone cannot be directly measured at the scene,firstly,a soft sensor model,adopting support vector machine optimized by improved particle swarm optimization,is established to predict the content of f-CaO,and then predict the temperature of burning zone,combining empirical relationship between f-CaO content and temperature of burning zone.In the temperature control system of the burning zone,the model predictive controller(DMC)is used to regulate the coal injection volume of the kiln head.When abnormal conditions occur,field operators adjust coal feed rate and kiln speed to suppress the strong interference in production with expert experience.Finally,this paper designs a computer control system for cement rotary kiln calcining on ECS-100 control system and APC-Suite software to realize the real-time monitoring of main parameters of cement clinker firing process.