水泥分解炉标准工况模板及建模研究

发布时间：2018-01-25 01:42

本文关键词： 分解炉工况模板回归分析灰色关联分析极限学习机 T-S模糊模型滑模控制　出处：《济南大学》2015年硕士论文　论文类型：学位论文

【摘要】：水泥分解炉作为新型干法窑外分解系统的核心设备,承担着燃烧、热传递和物料分解的任务。若分解炉出口温度太高,旋风预热器会发生结皮现象;相反,若分解炉出口温度过低,炉内的碳酸盐分解率会降低。因此,保持合理稳定的分解炉的出口温度是分解炉控制的关键所在。考虑到分解炉具有工况波动频繁、非线性、且延时较大等特点,直接建立模型以描述分解炉出口温度变化动态较为困难,为此,需建立完备的分解炉工况模板,而后根据不同的工况建立相应的分解炉数学模型,以便更加准确的实施分解炉出口温度控制。在水泥生产过程能效分析与优化控制理论方法研究(山东省自然科学基金重点项目:ZR2011FZ002)及新型干法水泥生产过程工况识别方法研究(山东省自然科学基金项目:ZR2010FM038)资助下,本文开展工作如下:(1)为精确控制分解炉出口温度,需建立准确的分解炉数学模型。考虑到分解炉工况波动频繁等因素,首先需要建立一个合理完备的分解炉标准工况模板,而后分工况建立分解炉数学模型。参考水泥工厂设计规范(GB 50295-2008),借鉴现场优秀操作人员工作经验,结合某5000t/d水泥厂生产线数据,分析得出喂煤量、生料下料量及三次风温度是影响分解炉出口温度的主要因素,绘制三维曲线,分析了各参变量间的关系,得到了合理的水泥分解炉标准工况模板,为后面建立基于标准工况模板的分解炉数学模型打下基础。(2)根据(1)中所建立的模板,选取喂煤量、生料下料量及三次风温度作为建模所需输入变量,分解炉出口温度作为输出变量,采用回归分析学习算法建立水泥分解炉出口温度在870-880℃时的数学模型。(3)考虑到现场应用,所建模型需具有在线校正能力,以及时准确反映分解炉温度变化动态。因此,在(2)中所述建模方法的基础之上,提出了一种基于灰色关联分析的分解炉出口温度在线切换建模方法。该方法能够实时计算喂煤量、生料下料量、三次风温度与分解炉出口温度之间的灰色关联度,选取关联度最大的变量作为建模的输入变量,建立了分解炉出口温度在870-880℃时的切换模型,并给出了在线模型校正方法。(4)为建立基于标准工况模板的分解炉数学模型,在(3)中所述建模方法的基础之上,提出了一种水泥分解炉出口温度在线T-S模糊建模方法。考虑到三次风温度主要受三次风量的影响,而分解炉在正常工作状态下,三次风阀门开度保持恒定,因此仅选用喂煤量及生料下料量作为建模所需输入变量,采用回归分析和极限学习机方法建立分解炉出口温度局部模型,根据水泥预分解工艺特点,划分隶属度曲线,给出了水泥分解炉出口温度在820-880℃时的T-S模糊模型。(5)为准确控制分解炉出口温度,需结合模型制定控制策略。根据上文所提建模方法,针对分解炉出口温度在840-860℃这一典型工况,建立了分解炉回归模型,并采用自适应趋近率求取最优控制量,设计了具有自适应能力的水泥分解炉滑模控制器。仿真结果证明了所建控制器的准确性。
[Abstract]:The core equipment of cement decomposing furnace as new dry kiln decomposition system, undertakes combustion, heat transfer and material decomposition task. If the decomposition furnace outlet temperature is too high, will happen cyclone preheater crust; on the contrary, if the decomposition furnace outlet temperature is too low, the furnace carbonate decomposition rate will be reduced. Therefore, to maintain export the temperature of decomposing furnace is the key to stable decomposition furnace control. Considering the condition of decomposition furnace with frequent fluctuations, nonlinear, large time delay and other characteristics, the direct model is established to describe the dynamic changes of the decomposition furnace temperature is difficult, need to establish a complete template decomposition furnace condition, and then the corresponding mathematical model of decomposition furnace set up according to different conditions, so that the implementation of more accurate decomposition furnace outlet temperature control. In the analysis of control theory and optimization method for the cement production process of energy (Natural Science of Shandong Province Science Fund Project: ZR2011FZ002) research and new dry cement production process condition recognition method (Natural Science Foundation of Shandong Province: ZR2010FM038) under the auspices of the work in this paper are as follows: (1) for the precise control of decomposition furnace outlet temperature of decomposing furnace, need to establish accurate mathematical model. Considering the decomposition furnace condition fluctuation first of all, the need to establish a reasonable and complete decomposition furnace standard condition template, then conditions to establish a decomposition furnace mathematical model. The reference code for design of cement plant (GB 50295-2008), from the experience of outstanding field operation personnel, combined with a 5000t/d cement production line data, analysis of coal feeding, raw material quantity and three air temperature is the main factor affecting the decomposition of furnace outlet temperature, 3D curve, analyzes the relationship between the variables, to obtain a reasonable Cement Decomposing Furnace standard conditions as template. The back up decomposition furnace mathematical model based on the template under standard conditions. (2) according to (1) in the template, select the coal feeding, raw material quantity and three wind temperature as the input variables needed for modeling, decomposition furnace outlet temperature as output variables, regression analysis was used to study the mathematical model of cement the decomposing furnace exit temperature at 870-880 DEG C to establish algorithm. (3) considering the field application, the model with online correction ability, and accurately reflect the dynamic changes of temperature of decomposing furnace. Therefore, in (2) on the basis of the modeling method, a grey relational analysis of decomposing furnace the outlet temperature of online modeling method based on switching. This method can real-time calculation of coal feeding, raw material quantity, grey correlation degree between the three air temperature and the decomposing furnace exit temperature, select the maximum correlation variables as input variables of the model, A switching model decomposing furnace exit temperature at 870-880 DEG C, and gives a model of online correction method. (4) to establish the mathematical model of decomposition furnace based on template in standard condition, (3) on the basis of the modeling methods, put forward a kind of cement decomposing furnace exit temperature online fuzzy modeling method T-S three. Considering the air temperature is mainly affected by the three volume, and decomposing furnace under normal working condition, the three valve opening is kept constant, so only use coal feeding and raw feed as needed for modeling the input variables by regression analysis and extreme learning machine method to establish the decomposing furnace exit temperature the local model, according to the process characteristics of cement precalciner, division membership curve, given the cement decomposing furnace exit temperature at 820-880 DEG C to T-S fuzzy model. (5) for accurate control of decomposing furnace exit temperature, should be combined with the formulation of control model Strategy. According to the modeling method mentioned above, the decomposing furnace exit temperature at 840-860 DEG. The typical working conditions, a decomposition furnace regression model, and the adaptive approach to obtain the optimal rate control, designed an adaptive sliding mode controller for Cement Decomposing Furnace. The simulation results prove that the accuracy of the controller.

【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ172.6

【相似文献】