基于关联规则挖掘的供热锅炉运行优化研究

发布时间：2018-06-11 11:16

本文选题：供热锅炉 + 数据挖掘　；参考：《大连海事大学》2017年硕士论文

【摘要】：传统的供热行业由于实际操作缺乏理论指导,管理上缺少有效地评价机制导致其能耗水平居高不下。所以研究供热锅炉的运行优化方式,为供热企业提供在实际生产过程当中有效地操作指导有着实际的意义。目前我国的供热企业主要使用链条式燃煤锅炉,而燃煤锅炉的燃烧过程极为复杂,各变量间相互耦合,并且具有大滞后、非线性等特点,所以要获取准确的数学模型非常困难。但随着信息技术的发展,在供热企业在生产过程中产生了大量的数据,且这些数据当中蕴含着大量与锅炉运行状态有关的信息。而由于企业不重视,同时又缺乏相关的统计学知识,这些数据没有没有效地利用。本文针对这一现状,在研究分析了锅炉运行机理后结合数据挖掘的手段对锅炉历史数据进行分析和关联规则的挖掘从而设计了一种针对锅炉操作变量的锅炉运行优化方法,主要内容如下:本文首先对燃煤热水锅炉的运行机理进行分析,确定将鼓风机转速、引风机转速、炉排转速、给煤机转速作为可操作变量。通过挖掘这4个操作变量与锅炉运行高效率点的关联规则,并根据关联规则调节操作变量达到对锅炉燃烧效率进行优化的目的。其次利用拉格朗日差值法等统计学方法对历史数据进行缺失数据的修补与异常值的剔除,并将处理后的历史数据进行离散化,选用Apriori法对历史数据进行关联规则的挖掘。对初步获取的关联规则进行合并与剔除形成关联规则库,并以此关联规则库作为锅炉运行优化指导的工具。然后详细说明了利用关联规则法对锅炉运行优化的具体步骤,根据锅炉运行的实时数据从关联规则库当中搜索符合当前锅炉运行状态的关联规则。以给煤机转速为基点,并根据关联规则中建议的操作变量取值区间调节锅炉的鼓风、引风、炉排等变量。最后将此优化方法在大连某公司的热水燃煤链条锅炉进行实际应用,验证了该方法的有效性。相比于传统的操作习惯,该操作方法可以明显的提升锅炉的燃烧效率。
[Abstract]:Because of the lack of theoretical guidance in the practical operation and the lack of effective evaluation mechanism in the traditional heating industry, the energy consumption level of the traditional heating industry remains high. Therefore, it is of practical significance to study the operation optimization mode of heating boiler, to provide effective operation guidance in the actual production process for heating enterprises. At present, chain type coal-fired boilers are mainly used in heating enterprises in our country, but the combustion process of coal-fired boilers is extremely complex, the variables are coupled with each other, and have the characteristics of large lag and nonlinear, so it is very difficult to obtain accurate mathematical models. However, with the development of information technology, a large number of data are produced in the production process of heating enterprises, and these data contain a large amount of information related to the operation state of the boiler. However, these data are not used effectively because the enterprises do not pay attention to and lack the relevant statistical knowledge at the same time. In view of this situation, after studying and analyzing the boiler operation mechanism and combining the means of data mining, this paper analyzes the boiler historical data and mining the association rules, and then designs a boiler operation optimization method aiming at the boiler operation variables. The main contents are as follows: firstly, the operating mechanism of coal-fired hot water boiler is analyzed, and the rotating speed of blower, induced fan, grate and coal feeder are determined as operable variables. By mining the four operating variables and regulating the operation variables according to the association rules, the boiler combustion efficiency can be optimized. Secondly, using the Lagrange difference method and other statistical methods to repair the missing data and eliminate the outliers, and discretize the processed historical data, Apriori method is used to mining the association rules of the historical data. The association rules are merged and eliminated to form the association rules base, and the association rules base is used as the guidance tool for boiler operation optimization. Then, the detailed steps of optimizing boiler operation by using association rule method are explained in detail. According to the real-time data of boiler operation, the association rules are searched from the association rules database to accord with the current boiler operation state. Based on the rotational speed of coal feeder, the variables such as blast, draft air and grate of boiler are adjusted according to the operation variables suggested in the association rules. Finally, the optimization method is applied to a hot water coal-fired chain boiler in Dalian Company, and the effectiveness of the method is verified. Compared with the traditional operation habits, this operation method can obviously improve the combustion efficiency of the boiler.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13;TU833.11

【参考文献】