水泥回转窑温度建模及控制研究

发布时间：2018-05-08 06:33

本文选题：水泥回转窑 + 烧成带温度　；参考：《天津理工大学》2017年硕士论文

【摘要】：水泥是建筑行业必不可少的材料,随着我国城镇化脚步的加快,水泥的需求量也逐步增加。水泥生产是一个热工过程,从原料的分解到烧结都需要很高的温度,回转窑烧成带的温度决定了水泥熟料的质量。因此,对水泥回转窑烧成带温度的准确获取和控制具有重要的现实意义和学术研究价值。由于回转窑系统存在非线性、大时滞、多变量等特点,传统的人工看火、比色高温计测温度、筒体扫描仪测温方式不能准确获取烧成带温度,进而影响烧成带温度控制。本文采用数据拟合的方法建立温度预测模型,并用模糊控制的方法对温度进行控制。本文首先简要介绍了新型干法水泥工业发展现状,总结了国内外学者在温度预测和控制方面取得的研究成果。通过分析水泥烧制的工艺流程,选择了九个与烧成带温度相关的变量,通过灰色关联度分析法计算了各变量和烧成带温度的关联程度,最终确定了七个用于建模的变量。在选定建模变量的基础上,根据回转窑系统自身的特点,选择了最小二乘支持向量机建立烧成带的温度模型,并用鲁棒性强、不易陷入局部最优的多种群遗传算法对最小二乘支持向量机的参数进行优化。通过仿真结果和实测数据的比较,验证了建模方法的有效性。由于受到物料成分波动的影响及外界环境的干扰,水泥回转窑系统的工况是时变的,为了更好的反映实际工况,提高预测的精度,研究提出了在线最小二乘支持向量机的温度预测算法。针对回转窑烧成带温度不容易实时检测的问题,选择易于在线检测的窑尾温度表征烧成带温度,作为模型的输出变量,该算法在线更新建模的数据,递推计算新模型参数,达到了在线预测温度的目的。温度控制系统选择喷煤量作为控制变量,被控量选择易于测量的窑尾温度。控制方法选择模糊控制,并设计了模糊控制器,建立了模糊规则库和查询表,仿真结果验证了控制算法的有效性。
[Abstract]:Cement is an indispensable material in the construction industry. With the acceleration of urbanization in China, the demand for cement is increasing gradually. Cement production is a thermal process, from the decomposition of raw materials to sintering requires a very high temperature, rotary kiln firing zone temperature determines the quality of cement clinker. Therefore, it has important practical significance and academic research value to accurately acquire and control the firing zone temperature of cement rotary kiln. Because the rotary kiln system has the characteristics of nonlinearity, large time delay and multivariable, the traditional manual fire monitoring, colorimetric high temperature measurement, and the tube scanner temperature measurement method can not accurately obtain the firing zone temperature, and then affect the firing zone temperature control. In this paper, the method of data fitting is used to establish the temperature prediction model, and the fuzzy control method is used to control the temperature. In this paper, the present situation of new dry cement industry is introduced briefly, and the research results of temperature prediction and control are summarized. Through analyzing the technological process of cement sintering, nine variables related to sintering zone temperature were selected, and the correlation degree between each variable and firing zone temperature was calculated by grey correlation degree analysis, and seven variables used for modeling were finally determined. On the basis of selecting modeling variables and according to the characteristics of rotary kiln system, the temperature model of firing band is established by using least square support vector machine, and the model is robust. Multi-population genetic algorithm, which is not easy to fall into local optimum, optimizes the parameters of least squares support vector machine (LS-SVM). The effectiveness of the modeling method is verified by comparing the simulation results with the measured data. Due to the influence of material composition fluctuation and external environment disturbance, the working conditions of cement rotary kiln system are time-varying. In order to better reflect the actual working conditions and improve the accuracy of prediction, A temperature prediction algorithm based on online least squares support vector machine (LS-SVM) is proposed. Aiming at the problem that the firing zone temperature of rotary kiln is not easy to detect in real time, the kiln tail temperature, which is easy to detect on line, is chosen as the output variable of the model. The algorithm updates the modeling data online and calculates the new model parameters by recursion. The purpose of on-line temperature prediction is achieved. The temperature control system selects the coal injection quantity as the control variable, and the controlled quantity selects the kiln tail temperature which is easy to measure. The fuzzy control method is selected and the fuzzy controller is designed. The fuzzy rule base and query table are established. The simulation results show the effectiveness of the control algorithm.
【学位授予单位】：天津理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TQ172.622

【相似文献】