资源描述：
《近红外光谱煤炭品质.pdf》由会员上传分享，免费在线阅读，更多相关内容在教育资源-天天文库。
1、第38卷第1期兰暋州暋理暋工暋大暋学暋学暋报Vol.38No.12012年2月JournalofLanzhouUniversityofTechnologyFeb.2012暋暋文章编号:1673灢5196(2012)01灢0059灢04近红外光谱灢系统聚类法快速测定煤炭品质李颖娜1,吴暋楠2,王宇亮2,徐志彬2(1.唐山学院环境与化学工程系,河北唐山暋063000;2.河北出入境检验检疫局京唐港办事处,河北唐山暋063000)摘要:采用偏最小二乘法对95个煤炭样品的近红外光谱数据进行处理,并提取主成分.将提取的主成分与煤炭的发热量、灰分
2、、挥发份、含硫量和全水分共同作为变量,进行系统聚类分析.将样品数据聚类为4组,同时剔除异常样本.对聚类后的各组数据采用多元散射校正、二阶导数、诺里斯导数平滑进行预处理,建立偏最小二乘定量分析模型.采用逐步筛选法,求得以发热量为变量的Bayes判别函数,交互验证结果表明判别函数稳定性良好.对未知样品发热量、灰分、挥发份、含硫量和全水分预测的决定系数分别达到0.992、0.927、0.938、0.778、0.978,说明模型预测性能良好.关键词:近红外光谱;煤炭;最小偏二乘;系统聚类;判别函数中图分类号:O657.33暋暋文献标识码:AF
3、astdeterminationofcoalqualitybyusingnear灢infraredspectroscopy灢hierarchicalclusteringmethod1,WUNan2,WANGYu灢liang2,XUZhi灢bin2LIYing灢na(1.DepartmentofEnvironmentalandChemicalEngineering,TangshanCollege,Tangshan暋063000,China;2.JingtangPortOfficeofHebeiEntry灢ExitInspectionan
4、dQuarantineBureau,Tangshan暋063000,China)Abstract:Partialleastsquare(PLS)methodwasusedtotreatthenear灢infraredspectradataof95coalsamplesandtheirprincipalcomponentswereextracted.Takingtheprincipalcomponentscombinedwiththecoalcalorificcapacity,ashcontent,volatility,sulphurc
5、ontent,andoverallmoisturecopacityasvaria灢bles,thesampleswerehierarchicallyclusteredintofourhomogeneousclustersand,meantion,theabnor灢malsampleswereexcluded.Thespectrumineveryclusterwerepreprocessedwithsecondderivativeandmultiplicativescatteringcorrection(MSC)andsmoothedw
6、ithNorrisderivativefilter,andthusthePLSquantitativeanalysismodelwasestablished.ThesuccesiveselectionmethodwasusedtoconstructaBayesdiscriminatefunctionwiththecalorificcapacilyasitsvariable.Theresultofinteractivevalidationindicatedthatthestabilizationofthisfunctionwasperf
7、ect.Thepredictiondecisioncoefficientofabove灢mentionedfivequantitiesofthecoalsampleswere0.992,0.927,0.938,0.778,0.978,showingthatthepredictiveperformanceofthemodelwasfine.Keywords:near灢infraredspectrum;coal;partialleastsquare;hierarchicalcluster;discriminatefunction暋暋煤炭的
8、发热量、灰分、挥发份和硫含量是评价煤建模往往不能收到很好的效果.因此按照样本的来炭品质的重要指标,传统化学分析方法耗时长,操作源地对大量样本进行细分后建模,不仅能够减少样[3]复杂.近红外光谱分析技术适合于多组分的定性和
