新型干法水泥游离氧化钙含量软测量算法研究

发布时间：2018-04-18 02:39

本文选题：水泥熟料游离氧化钙 + 软测量　；参考：《昆明理工大学》2015年硕士论文

【摘要】：水泥是世界上最重要的建筑材料,是三大建材材料支柱之一。然而,我国的水泥产业同时也存在诸如水泥质量差,耗能高,污染重等缺点。因此,为确保水泥的安全生产,生产过程中的在线监控和质量控制是十分必要的。水泥熟料的品质决定着整个水泥的产品质量,而水泥熟料中游离氧化钙(f-CaO)的含量可以最直接、最快速地反映水泥熟料的质量好坏。为了提高产品的经济价值,保证水泥熟料的质量品质,就要实时分析熟料中游离氧化钙的含量。然而,由于水泥生产过程复杂多变,对实际生产中的f-CaO含量进行在线分析十分困难,目前主要依靠人工化验分析得到。而人工化验分析f-CaO含量,存在测定时间过长、滴定终点难以判断、不能及时反映其变化等问题,远远不能满足现代化大规模水泥熟料生产质量控制的要求。软测量技术是利用相关的可测变量(辅助变量),建立待测变量与辅助变量之间的数学模型,以实现难测变量的在线预测。因此,针对水泥熟料中f-CaO含量难以在线分析的现状,本论文首先在深入分析云南某水泥厂2000t/d新型干法水泥生产工艺流程及特点的基础上,采集了其DCS系统中记录的在线可测数据及台帐化验数据,并进行了数据处理与分析；其次,找出了影响游离氧化钙含量的主要因素(即生料的三率值、分解炉喂煤量、回转窑喂煤量、喂料量、回转窑温度),并利用多元回归法(MLS)、主元回归及主元分析法(PCA-PCR)、偏最小二乘法(PLS)以及径向基(RBF)、支持向量机(SVM)等多种软测量方法,分别建立了f-CaO含量的软测量模型；最后,通过不同软测量模型的仿真验证与对比分析研究,确定出基于支持向量机的f-CaO含量软测量模型具有最佳的预测效果。本论文的研究可为水泥熟料的产品质量在线监测与控制奠定理论基础,以确保产品质量、提高经济效益并降低生产成本。
[Abstract]:Cement is the world's most important building materials, is one of the three pillars of building materials.However, China's cement industry also has shortcomings such as poor cement quality, high energy consumption, heavy pollution and so on.Therefore, in order to ensure the safety of cement production, online monitoring and quality control in the production process is very necessary.The quality of cement clinker determines the product quality of the whole cement, and the content of free calcium oxide f-CaO in cement clinker can reflect the quality of cement clinker most directly and quickly.In order to improve the economic value of products and ensure the quality and quality of cement clinker, it is necessary to analyze the content of free calcium oxide in clinker in real time.However, due to the complex and changeable process of cement production, it is very difficult to analyze the content of f-CaO in actual production on line.However, the determination time is too long, the titration end point is difficult to judge, and the change of f-CaO content can not be reflected in time. It is far from meeting the requirement of modern large-scale cement clinker production quality control.The soft sensing technique is to establish the mathematical model between the variables to be tested and the auxiliary variables by using the relevant measurable variables (auxiliary variables) in order to realize the on-line prediction of the difficult variables.Therefore, in view of the present situation that the content of f-CaO in cement clinker is difficult to be analyzed on line, this paper firstly analyzes the technological process and characteristics of 2000t/d dry cement production in a cement plant in Yunnan province.The on-line measurable data and laboratory data recorded in its DCS system were collected, and the data were processed and analyzed. Secondly, the main factors affecting the content of free calcium oxide were found out, that is, the three rate value of raw meal, the coal feed amount of calciner.Various soft sensing methods, such as coal feed, feed quantity, temperature of rotary kiln, multiple regression method, principal component regression and principal component analysis (PCA-PCRN), partial least square method (PLS), radial basis function (RBF) and support vector machine (SVM) are used.The soft sensor models of f-CaO content are established respectively. Finally, through the simulation verification and comparative analysis of different soft sensor models, it is determined that the soft sensor model based on support vector machine has the best prediction effect.The research in this paper can lay a theoretical foundation for the on-line monitoring and control of the product quality of cement clinker in order to ensure the product quality, improve the economic benefit and reduce the production cost.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ172.18

【参考文献】