锅炉氮氧化物排放量检测中PLS的应用改进  

The application improvement of PLS in boiler NO_x emissions testing

作　　者：董增寿[1] 刘明君[1] 

机构地区：[1]太原科技大学电子信息工程学院,太原山西030024

出　　处：《山东大学学报（工学版）》2010年第1期126-128,138,共4页Journal of Shandong University（Engineering Science）

基　　金：山西省自然科学基金资助项目(2007011048)

摘　　要：通过确定与氮氧化物生成量密切相关的锅炉运行参数,利用改进的偏最小二乘法(partial least squares,PLS)计算模型对锅炉煤燃烧过程中氮氧化物排放量进行预测。PLS的改进是在随机矩阵法均化校准模型基础上,结合影响氮氧化物排放量的各个因素以及排放量中氮含量的单变量输出,对PLS模型改进。利用PLS改进的计算模型对氮氧化物排放量进行预测,并同实测值进行对比,结果表明:该方法对预测氮氧化物排放量的准确度和计算速度较常规的PLS均有较大提高。该方法为实现数据采集程序计算一体化提供了一种手段。By identifying boiler operating parameters closely related to nitrogen oxides emissions, the improvement of PLS was used to predict the nitrogen oxides emissions in the boiler coal combustion process. Based on calibration models uniformed by the random matrix and combined with several factors influenced by nitrogen oxides emissions, and variable single-output of the emis- sions of nitrogen content, the PLS model was improved. The improved calculation model was used to test nitrogen oxides emis- sions. The result showed that this method was more accurate and faster than conventional PLS when used to predict the emission of nitrogen oxides. This method could provide a means for the integration of data calculation and process collection.

关 键 词：偏最小二乘法 随机矩阵 模型 氮氧化物 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]