基于LabVIEW的水泥熟料游离氧化钙BP软测量及质量统计监控研究

发布时间：2018-02-24 08:14

本文关键词： 游离氧化钙软测量 BP神经网络 ADO MATLAB Script LabVIEW DSC　出处：《昆明理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：新型干法水泥生产是我国乃至世界水泥生产的主要技术之一,其生产过程复杂、连续性强、相互影响、相互制约,尤其在窑内煅烧工序中必须保证水泥熟料游离氧化钙(f-CaO)含量维持在合理的范围内。随着游离氧化钙含量增加,熟料的强度和安定性都会大幅下降；而游离氧化钙含量一旦低至0.5%以下时,熟料甚至会进入死烧状态。因此,熟料游离氧化钙含量控制的好坏直接关系到水泥产品的质量。在实际生产中,影响游离氧化钙含量的因素较多,包括投料量,头煤量,尾煤量,三率值等,目前主要依靠人工每小时化验一次得到熟料f-CaO含量。而要保证水泥熟料的质量,熟料f-CaO的含量应维持在工艺需求的0.5%～2.0%之间。针对f-CaO含量难以在线测量,本论文首先利用软测量技术中的BP神经网络建模方法,建立了f-CaO含量软测量模型并验证了模型的有效性和准确性并进行了游离氧化钙含量软测量模型的短期校正研究；其次,为了便于对水泥熟料性能进行分析与监控,本论文以LabVIEW作为开发平台,开发了基于SQL2008R2的新型干法水泥生产过程数据库的友好人机监控界面。本论文的核心工作是利用MATLAB Script节点实现了MATLAB游离氧化钙含量的BP神经网络软预测；利用LabVIEW DSC模块设计了监控画面并采用均值—标准差控制图来实现新型干法水泥熟料游离氧化钙质量统计分析与预警,以指导操作人员判断生产过程中熟料f-CaO含量是否达标,进而采取有效的控制手段；利用ADO技术实现了对数据库的访问与数据存储,可方便地查询、显示f-CaO含量的软预测值及其变化趋势。本论文的特点是充分利用了LabVIEW模块化、图形化、每个子程序独立调用的编程特点,以菜单方式实现了基于MATLAB Script节点、LabVIEW DSC统计分析模块及ADO数据库访问技术的新型干法水泥生产水泥游离氧化钙含量的软预测与质量监控,完成了多个系统功能的集成开发。本论文的工作可为新型干法水泥生产过程熟料游离氧化钙含量的在线监测提供一种有效途径并奠定了一定的理论基础。
[Abstract]:The new dry process cement production is one of the main technologies of cement production in our country and even in the world. Its production process is complex, continuous, mutual influence and mutual restriction. Especially in the kiln calcination process, the content of free calcium oxide (f-CaO) in cement clinker must be kept within a reasonable range. With the increase of free calcium oxide content, the strength and stability of clinker will decrease greatly. However, once the content of free calcium oxide is below 0.5%, the clinker will even enter into a dead state. Therefore, the quality of cement products is directly related to the control of free calcium oxide content in clinker. There are many factors that influence the content of free calcium oxide, including the quantity of feed, the quantity of coal, the quantity of tail coal, the value of three-rate, etc. At present, the f-CaO content of clinker is obtained mainly by manual test once an hour, and the quality of cement clinker should be guaranteed. The content of f-CaO in clinker should be kept between 0.5% and 2.0% of the technological requirement. The f-CaO soft sensor model is established, and the validity and accuracy of the model are verified, and the short-term correction research of the free calcium oxide soft sensor model is carried out. Secondly, in order to facilitate the analysis and monitoring of cement clinker performance, This paper takes LabVIEW as the development platform, The friendly man-machine monitoring interface of a new dry cement production process database based on SQL2008R2 is developed. The core work of this paper is to use MATLAB Script node to realize the BP neural network soft prediction of MATLAB free calcium oxide content. Using LabVIEW DSC module, the monitoring screen is designed and the mean value standard deviation control chart is used to realize the statistical analysis and early warning of the free calcium oxide quality of the new dry cement clinker, so as to guide the operators to judge whether the f-CaO content of clinker is up to standard in the production process. Then the effective control means are adopted, the database access and data storage are realized by using ADO technology, and the soft prediction value of f-CaO content and its changing trend can be easily queried. The characteristic of this paper is that the modularization of LabVIEW is fully utilized. Graphical, the programming characteristic of each subroutine calling independently, Based on MATLAB Script node LabVIEW DSC statistical analysis module and ADO database access technology, the soft prediction and quality control of free calcium oxide content in new dry cement production are realized by menu method. The work of this paper can provide an effective way for on-line monitoring of free calcium oxide content of clinker in the process of new dry cement production and lay a certain theoretical foundation.
【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】：TQ172.16;TP311.52

【参考文献】