基于机器学习的煤质近红外光谱分析
|
摘要
为充分有效地利用煤炭资源，须及时掌握煤质的变化规律。受测量方法和相关技术限制，传统的煤质分析技术已不能满足煤炭生产、加工和利用等过程的要求。煤质近红外光谱分析技术是一种新兴的煤质快速检测方法，可实现煤质全元素的快速在线分析。但由于该技术是一种间接分析方法，预测结果的准确性主要依赖于建模数据及建模方法。鉴于此，针对煤样光谱数据存在的不稳定因素多、维数高、特征变异范围广等问题，研究建模样本的优化选择、光谱数据的恢复与压缩处理，以及定量分析模型的构建。
针对建模样本的优化选择，首先，研究同一煤样在不同粒度等级下的煤样光谱数据对分析模型预测精度的影响，采集0.2mm、1mm、3mm和13mm等4种粒度下的光谱数据，利用统一的测试平台与评估函数给出各种光谱数据的准确度对比；然后，提出基于迭代裁剪法的异常样本剔除，以相异性测度作为判定指标，结合光谱数据的特点给出相应的裁剪准则；最后，利用相关性测度制定判别准则，提出基于并行最小二乘回归估计法的争议样本判别。
针对光谱数据的恢复处理，首先分析常用光谱恢复方法的适用性；然后，提出基于拟线性局部加权法的光谱散射校正模型，以及基于粗糙惩罚法的光谱优化平滑模式。通过引入拟线性函数与核函数、粗糙惩罚项，对原光谱与“理想”光谱在各波长点处的依赖关系进行更精确的描述，最大限度地消除散射干扰和随机噪声，使得恢复处理后的光谱更加接近真实光谱。
针对光谱数据的压缩处理，首先，给出几种以波长点或谱区为单位的特征选择方法，包括重要性排序的向前选择法和基于遗传算法的谱区优化组合；然后，针对特征选择方法降维效果不显著的缺陷，在线性主成分分析方法的基础上，提出基于核主成分分析方法的光谱特征提取，该方法不仅改善了线性主成分分析方法的缺陷，还能实现光谱数据的高效压缩。
利用统一的测试平台和评估参数，给出建模样本优化、光谱数据恢复与压缩等方法，对不同质量(信噪比)光谱数据集的分析处理效果，以全面验证各理论与方法的泛化性和有效性。
针对煤质近红外光谱定量分析模型的构建，通过借鉴集成学习的思想，综合多种神经网络方法的优势，提出基于集成神经网络的定量分析模型：首先，利用SOM神经网络模型将全部建模数据集分为若干类子集，将其变异范围进行细化，以解决煤炭光谱特征信息分散的问题；然后，建立基于RBF和BP神经网络的定量分析子模型，并将各子模型的输出信息进行融合，以提高预测结果的可靠性。此外，为了减少参数设置对模型预测性能的影响，分别利用先验知识、交叉验证和遗传算法等方法，对分类个数、扩展常数和权值阈值等参数进行优化搜索。
本文的研究综合了煤化学、光谱学、机器学习和人工智能等多门学科，产生的研究成果能够丰富光谱数据处理与回归预测的相关理论，并可有效提高煤质近红外光谱分析模型的预测能力。同时，对近红外光谱分析技术在煤质分析中的应用和推广，具有重要的理论意义和实际应用价值。
In order to effectively utilize coal resources, the change rule of coal qualitiesmust be grasped timely. The traditional coal analytical methods, restricted by itscorresponding measuring methods and techniques, can hardly satisfy therequirements on coal producing, processing and utilizing. The near-infraredspectroscopy (NIRS) can rapidly realize the online analysis for coal qualities.However, as the technology is an indirect analytical method for coal detection, itsprediction accuracy mainly depends on the modeling data and methods. In spectraldata of coal samples, there are numerous destabilizing factors, high dimensions andextensive characteristic variation. To solve the above problems, this paper studies thetheories and methods for optimization of the modeling samples, recovery andcompression of spectral data and establishing of quantitative analysis model.
Aiming at the optimization of modeling samples, firstly, this paper studies theeffect of coal particle size on the analytical model in NIRS, collects the spectral datafrom the same coal sample with four different particle sizes, i.e.,0.2mm,1mm,3mm and13mm, and adopts unified test platform and evaluation function to figureout the accuracy of each spectral data set. Then, the following methods are putforward: elimination of abnormal spectra based on iteration clipping method;discrimination of dispute samples based on parallel least-square regressionestimation method, which makes the measure of diversity and relativity as thedistinguishing factors, and figures out the corresponding distinguishing criterionaccording to spectral characteristics.
Aiming at the recovery processing of spectral data, firstly, an applicabilityanalysis is given about commonly-used spectral recovery methods. Then, this paperputs forward the quasi-linear local weighting method for spectral scatter correction,and the spectral optimization smoothing mode basing on roughness penalty rule. Bymeans of introducing quasi-linear function, kernel function and roughness penaltyterms, the dependencies between the original spectra and the ideal spectra at eachwavelength is described more accurately and meticulously, which maximumlyeliminates the scattering influence and random noise and makes the recoveredspectra closer to the real spectra.
In term of the compression processing of spectra data, firstly, several featureselection methods having wavelength point or spectral range as the unit have been given, which include forward selection method with importance ranking ofwavelength point or spectral region, and genetic algorithm to obtain the optimizationcombination of spectral feature regions. Then, aiming at the effect of unobviousdimension reduction of feature selection method, the spectral feature extractionmethod is put forward basing on the kernel principle component analysis, which cannot only improve the defects of linear principal component analysis method, but alsocompress the spectral data with high-effective.
With the use of the unified test platform and evaluation parameter, the analyticaleffects of spectral data sets with different quality (noise ratio) on modeling sampleoptimization and spectral data recovery and compression, are analyzed in order tosystematically verify the generalization and effectiveness of all the proposed theoriesand methods in this paper.
Aiming at structuring quantitative analysis model of coal qualities in NIRS, bydrawing the idea from the integrated study, a quantitative analysis model based onintegrated neural networks is presented with the synthesizing advantages of varietyneural network. Firstly, with the help of SOM neural network, all the modeling datasets have been divided into several subsets, which can detail the range of variation,in order to solve the problem of decentralized characteristic information of modelingdata set; then, the quantitative analytical sub-models are constructed based on RBFand BP neural network, and the output information of all the sub-models areintegrated to improve the reliability of the prediction results. In addition, to reducethe parameter setting influence on the performance predicting of each sub-models,this paper utilizes the methods of priori knowledge, cross validation and geneticalgorithm to search for the optimal parameters, such as the classification number,extending constant, weights and thresholds etc.
In this paper, many subjects are intergrated for analyzing coal qualities withNIRS, including coal chemistry, spectroscopy, machine learning and artificialintelligence and so on. The research work will enrich the related theories aboutspectral data processing and regression prediction, and improve the predictive abilityof analytical model effectively. Besides, the finding of this paper has both importanttheoretical significance and practical application value.
针对建模样本的优化选择，首先，研究同一煤样在不同粒度等级下的煤样光谱数据对分析模型预测精度的影响，采集0.2mm、1mm、3mm和13mm等4种粒度下的光谱数据，利用统一的测试平台与评估函数给出各种光谱数据的准确度对比；然后，提出基于迭代裁剪法的异常样本剔除，以相异性测度作为判定指标，结合光谱数据的特点给出相应的裁剪准则；最后，利用相关性测度制定判别准则，提出基于并行最小二乘回归估计法的争议样本判别。
针对光谱数据的恢复处理，首先分析常用光谱恢复方法的适用性；然后，提出基于拟线性局部加权法的光谱散射校正模型，以及基于粗糙惩罚法的光谱优化平滑模式。通过引入拟线性函数与核函数、粗糙惩罚项，对原光谱与“理想”光谱在各波长点处的依赖关系进行更精确的描述，最大限度地消除散射干扰和随机噪声，使得恢复处理后的光谱更加接近真实光谱。
针对光谱数据的压缩处理，首先，给出几种以波长点或谱区为单位的特征选择方法，包括重要性排序的向前选择法和基于遗传算法的谱区优化组合；然后，针对特征选择方法降维效果不显著的缺陷，在线性主成分分析方法的基础上，提出基于核主成分分析方法的光谱特征提取，该方法不仅改善了线性主成分分析方法的缺陷，还能实现光谱数据的高效压缩。
利用统一的测试平台和评估参数，给出建模样本优化、光谱数据恢复与压缩等方法，对不同质量(信噪比)光谱数据集的分析处理效果，以全面验证各理论与方法的泛化性和有效性。
针对煤质近红外光谱定量分析模型的构建，通过借鉴集成学习的思想，综合多种神经网络方法的优势，提出基于集成神经网络的定量分析模型：首先，利用SOM神经网络模型将全部建模数据集分为若干类子集，将其变异范围进行细化，以解决煤炭光谱特征信息分散的问题；然后，建立基于RBF和BP神经网络的定量分析子模型，并将各子模型的输出信息进行融合，以提高预测结果的可靠性。此外，为了减少参数设置对模型预测性能的影响，分别利用先验知识、交叉验证和遗传算法等方法，对分类个数、扩展常数和权值阈值等参数进行优化搜索。
本文的研究综合了煤化学、光谱学、机器学习和人工智能等多门学科，产生的研究成果能够丰富光谱数据处理与回归预测的相关理论，并可有效提高煤质近红外光谱分析模型的预测能力。同时，对近红外光谱分析技术在煤质分析中的应用和推广，具有重要的理论意义和实际应用价值。
In order to effectively utilize coal resources, the change rule of coal qualitiesmust be grasped timely. The traditional coal analytical methods, restricted by itscorresponding measuring methods and techniques, can hardly satisfy therequirements on coal producing, processing and utilizing. The near-infraredspectroscopy (NIRS) can rapidly realize the online analysis for coal qualities.However, as the technology is an indirect analytical method for coal detection, itsprediction accuracy mainly depends on the modeling data and methods. In spectraldata of coal samples, there are numerous destabilizing factors, high dimensions andextensive characteristic variation. To solve the above problems, this paper studies thetheories and methods for optimization of the modeling samples, recovery andcompression of spectral data and establishing of quantitative analysis model.
Aiming at the optimization of modeling samples, firstly, this paper studies theeffect of coal particle size on the analytical model in NIRS, collects the spectral datafrom the same coal sample with four different particle sizes, i.e.,0.2mm,1mm,3mm and13mm, and adopts unified test platform and evaluation function to figureout the accuracy of each spectral data set. Then, the following methods are putforward: elimination of abnormal spectra based on iteration clipping method;discrimination of dispute samples based on parallel least-square regressionestimation method, which makes the measure of diversity and relativity as thedistinguishing factors, and figures out the corresponding distinguishing criterionaccording to spectral characteristics.
Aiming at the recovery processing of spectral data, firstly, an applicabilityanalysis is given about commonly-used spectral recovery methods. Then, this paperputs forward the quasi-linear local weighting method for spectral scatter correction,and the spectral optimization smoothing mode basing on roughness penalty rule. Bymeans of introducing quasi-linear function, kernel function and roughness penaltyterms, the dependencies between the original spectra and the ideal spectra at eachwavelength is described more accurately and meticulously, which maximumlyeliminates the scattering influence and random noise and makes the recoveredspectra closer to the real spectra.
In term of the compression processing of spectra data, firstly, several featureselection methods having wavelength point or spectral range as the unit have been given, which include forward selection method with importance ranking ofwavelength point or spectral region, and genetic algorithm to obtain the optimizationcombination of spectral feature regions. Then, aiming at the effect of unobviousdimension reduction of feature selection method, the spectral feature extractionmethod is put forward basing on the kernel principle component analysis, which cannot only improve the defects of linear principal component analysis method, but alsocompress the spectral data with high-effective.
With the use of the unified test platform and evaluation parameter, the analyticaleffects of spectral data sets with different quality (noise ratio) on modeling sampleoptimization and spectral data recovery and compression, are analyzed in order tosystematically verify the generalization and effectiveness of all the proposed theoriesand methods in this paper.
Aiming at structuring quantitative analysis model of coal qualities in NIRS, bydrawing the idea from the integrated study, a quantitative analysis model based onintegrated neural networks is presented with the synthesizing advantages of varietyneural network. Firstly, with the help of SOM neural network, all the modeling datasets have been divided into several subsets, which can detail the range of variation,in order to solve the problem of decentralized characteristic information of modelingdata set; then, the quantitative analytical sub-models are constructed based on RBFand BP neural network, and the output information of all the sub-models areintegrated to improve the reliability of the prediction results. In addition, to reducethe parameter setting influence on the performance predicting of each sub-models,this paper utilizes the methods of priori knowledge, cross validation and geneticalgorithm to search for the optimal parameters, such as the classification number,extending constant, weights and thresholds etc.
In this paper, many subjects are intergrated for analyzing coal qualities withNIRS, including coal chemistry, spectroscopy, machine learning and artificialintelligence and so on. The research work will enrich the related theories aboutspectral data processing and regression prediction, and improve the predictive abilityof analytical model effectively. Besides, the finding of this paper has both importanttheoretical significance and practical application value.
引文
[1]刘炯天.关于我国煤炭能源低碳发展的思考[J].中国矿业大学学报(社会科学版),2011,1:5-12.
[2]何选明.煤化学(第2版)[M].北京:冶金工业出版社,2010,5.
[3]邢志良.煤炭生产中全过程质量控制方法研究[J].中国矿业,2010,19(3):44-47.
[4]滕吉文,张雪梅,杨辉.中国主体能源——煤炭的第二深度空间勘探、开发和高效利用[J].地球物理学进展,2008,23(4):972-992.
[5]陈建国,张庆福,马玖坤,张鹤鹏.加强煤炭质量管理提高商品煤销售效益[J].煤炭技术,2005,24(2):119-120.
[6]张双全,吴国光.煤化学[M].徐州:中国矿业大学出版社,2004.
[7]于实,李丰田.煤质检测分析新技术新方法与化验结果的审查计算实用手册[M].北京:当代中国音像出版社,2005,5.
[8]施玉英.煤炭分析试验仪器设备的使用与维修[M].北京:煤炭工业出版社,2007,8.
[9]陆婉珍,袁洪福,楮小立.近红外光谱仪器[M].北京:化学工业出版社,2010,2.
[10] Yasukuni R., Boubekri R., Grand J., Felidj N., Maurel F., Perrier A., Metivier R., NakataniK., Yu P., Aubard J.. Specific and nondestructive detection of different diarylethene isomersby NIR-SERS[J]. Journal of Physical Chemistry C,2012,116(30):16063-16069.
[11] Bahareh J., Saeid M., Ezzedin M., Hassan G.. Reflectance Vis/NIR spectroscopy fornondestructive taste characterization of Valencia oranges[J]. Computers&Electronics inAgriculture,2012,85:64-69.
[12] Ni Yong, Shao Yongni, He Yong. Fast nondestructive measurement and discrimination oftablet based on Vis/NIR spectroscopy and chemometrics[C]. Proceedings of The3rdInternational Conference on Measuring Technology&Mechatronics Automation,2011,48-49:506-510.
[13] Raúl G., Antonio J.S., J.G., Eugenio I., Ana F., Jose M.B.. Nondestructive assessment offreshness in packaged sliced chicken breasts using SW-NIR spectroscopy[J]. Food ResearchInternational,2010,44(1):331-337.
[14] Tripathi S., Patel K.G., Bafna A.M.. Nondestructive determination of curcuminoids fromturmeric powder using FT-NIR[J]. Journal of Food Science&Technology-Mysore,2010,47(6):678-681
[15]谢晓霞,付利俊,庞文娟.影响煤工业分析特性指标检测的关键步骤[J].煤质技术,2010,6:33-34.
[16] Gu Deshan, Cheng Daowen, Yin Jie, Qiao Shuang, Liu Linmao, Zhang Xuesong. Rapidanalysis of coal properties[J]. IEEE Instrumentation&Measurement Magazine,2005,12:22-27.
[17] Xia Wencheng, Yang Jianguo, Liang Chuan. Effect of microwave pretreatment on oxidizedcoal flotation[J]. Powder Technology,2013,233:186-189
[18] Fu Huilin, Shen Zhongli. Microwave heating-based analysis method for the coal moisturecontent[J]. Energy Education Science&Technology Part A-energy Science&Research,2011,28(1):293-300.
[19] Arash T., Yu J.L., Li X.Ch., Meesri C.. Experimental study on microwave drying of Chineseand Indonesian low-rank coals[J]. Fuel Processing Technology,2011,92(10):1821-1829.
[20] Ozbayoglu G., Depci T., Ataman N.. Effect of microwave radiation on coal flotation[J].Energy Sources Part A-Recovery Utilization&Environmental Effects,2009,31(6):492-499.
[21] Liu Haiyu, Tan Houzhang, Gao Qiang, Hu Hongli, Xu Tongmo. Experimental study of fastmeasuring moisture content of coal by microwave[C]. Proceedings of The7th InternationalSymposium on Test&Measurement,2007,1-7:1937-1941.
[22]王辉东.中子活化煤质在线分析及β辐射伏特核电池能量输运研究[D].吉林大学,2012.
[23] He L.L., Melnichenko Y.B., Mastalerz M., Sakurovs R., Radlinski A.P., Blach Tomas. PoreAccessibility by methane and carbon dioxide in coal as determined by neutron scattering[J].Energy&Fuels,2012,26(3):1975-1983.
[24] Rey R.M.A., Alonso S.T., Castro G.M.P.. Neutron activation of deferred gamma rays steps todetermine the usefulness of this technique in analyzing elements of a mineral sample[C].Proceedings of The5th Meeting of the Spanish Neutron Scattering Association,2011,325:12025-12029.
[25] Lim C.S., Abernethy D.A.. On-line coal analysis using fast neutron-induced gamma-rays[J].Applied Radiation&Isotopes,2005,63(5-6):697-704.
[26] Akkaya, A.V.. Predicting coal heating values using proximate analysis via a neural networkapproach[J]. Energy Sources Part A-Recovery Utilization&Environmental Effects,2013,35(3):253-260.
[27]王建军,王世营,雷萌.粒子群优化算法在煤炭发热量预测模型中的应用[J].工矿自动化,2012,5:50-53.
[28] Chelgani S.C., Hart B., Grady W.C., Hower J.C.. Study relationship between inorganic andorganic coal analysis with gross calorific value by multiple regression and ANFIS[J].International Journal of Coal Preparation&Utilization,2011,31(1):9-19.
[29] Yao Wanye, Su Ling, Yin Shi. The application of BP neural network in coal analysis[C].Proceedings of The2011International Conference on Machine Learning&Cybernetics,2011,7:795-798.
[30] Jorjani E., Poorali H.A., Sam A., Chelgani S. Ch., Mesroghli Sh., Shayestehfar M.R..Prediction of coal response to froth flotation based on coal analysis using regression andartificial neural network[J]. Minerals Engineering,2009,22(11):970-976.
[31]周翠红,路迈西.线性回归与人工神经网络预测煤炭发热量[J].煤炭科学技术,2009,37(2):117-120.
[32]江文豪,韦红旗,屈天章,朱锋.基于遗传算法优化参数的支持向量机燃煤发热量预测[J].热力发电,2011,40(3):14-19.
[33] Burgess C., Hammond J.. Exploring the "forgotten" region-an update on NIRspectrometry[J]. Spectroscopy Europe,2009,21(2):28-31.
[34]严衍禄,赵龙莲,韩东海,杨曙明.近红外光谱分析基础及应用[M].北京:中国轻工业出版社,2005,1.
[35] Blanco M., Villarroya I.. NIR spectroscopy: a rapid-response analytical tool[J]. Trends inAnalytical Chemistry,2002,21(4):240-250.
[36]徐广通,袁洪福,陆婉珍.近红外光谱定量校正模型适用性研究[J].光谱学与光谱分析2001,21(4):459-463.
[37]徐广通,袁洪福,陆婉珍.现代近红外光谱技术及应用进展[J].光谱学与光谱分析,2000,20(2):134-142.
[38] Salgo A., Gergely S.. Analysis of wheat grain development using NIR spectroscopy[J].Journal of Cereal Science,2011,56(1):31-38.
[39] Shetty N., Gislum R., Jensen A.M.D., Boelt B.. Development of NIR calibration models toassess year-to-year variation in total non-structural carbohydrates in grasses using PLSR[J].Chemometrics&Intelligent Laboratory Systems,2012,111(1):34-38.
[40] Lin Hao, Zhao Jiewen, Sun Li, Chen Quansheng, Zhou Fang. Freshness measurement ofeggs using near infrared spectroscopy and multivariate data analysis[J]. Innovative FoodScience&Emerging Technologies,2011,12(2):182-186.
[41] Lourdes S.Ch., Vincent B., Ouissam A., Francisco P.R.. On-line analysis of intact olive fruitsby vis-NIR spectroscopy: Optimisation of the acquisition parameters[J]. Journal of FoodEngineering,2012,112(3):152-157.
[42] Kurihara K., Kawaguchi H., Obata T., Ito H., Sakatani K., Okada E.. The influence of frontalsinus in brain activation measurements by near-infrared spectroscopy analyzed by realistichead models[J]. Biomedical Optics Express,2012,3(9):2121-2130.
[43] De Bleye C., Chavez P.F., Mantanus J., Marini R., Hubert Ph., Rozet E., Ziemons E.. Criticalreview of near-infrared spectroscopic methods validations in pharmaceutical applications[J].Journal of Pharmaceutical&Biomedical Analysis,2012,69(SI):125-132.
[44] Prieto N., Dugan M.E.R., Lopez C.O., McAllister, T.A., Aalhus, J.L., Uttaro, B.. Nearinfrared reflectance spectroscopy predicts the content of polyunsaturated fatty acids andbiohydrogenation products in the subcutaneous fat of beef cows fed flaxseed[J]. MeatScience,2012,90(1):43-51.
[45] Kodzue K., Hiroyuki M., Mari M., Natsuki H., Hideyasu K., Hiroshi K., Wang P.Y., OsamuI., Hiroshi K., Roumiana T.. Near infrared spectroscopy of urine proves useful for estimatingovulation in giant panda (ailuropoda melanoleuca)[J]. Analytical Methods,2010,2:1671-1675.
[46]孔翠萍,褚小立,杜泽学,陆婉珍.近红外光谱方法预测生物柴油主要成分[J].分析化学,2010,38(6):805-810.
[47] Kaihara M., Takahashi T., Akazawa T., Sato T., Takahashi S.. Application of near infraredspectroscopy to rapid analysis of coals[J]. Spectroscopy Letters,2002,35(3):369-376.
[48] Andres J.M., Bona M.T.. Analysis of coal by diffuse reflectance near-infrared spectroscopy[J]. Analytica Chimica Acta,2005,535:123-132.
[49] Andres J.M., Bona M.T.. Coal analysis by diffuse reflectance near-infrared spectroscopy:Hierarchical cluster and linear discrimination analysis[J]. Talanta,2007,72:1423-1431.
[50] Andres J.M., Bona M.T.. ASTM clustering for improving coal analysis by near-infraredspectroscopy[J]. Talanta,2006,70:711-719.
[51] Dong Won Kim, Jong Min Lee, Jae Sung Kim. Application of near infrared diffusereflectance spectroscopy for on-line measurement of coal properties[J]. Korean Journal ofChemical Engineering,2009,26(2):489-495.
[52]武中臣,熊智新,王海东,张淑宁,张鹏彦,贾滨,贺守波.褐煤品质的傅里叶变换近红外光谱定量分析[J].光谱实验室,2008,25(5):819-823.
[53]丁仁杰,张笑.基于近红外光谱分析的电厂煤质分类快速检测方法[J].电站系统工程,2007,23(1):32-34.
[54]邬蓓蕾,林振兴,王群威,林力.傅立叶变换近红外光谱定量分析煤炭挥发分[J].岩矿测试,2006,25(2):133-136.
[55]邬蓓蕾,林振兴,王群威.傅立叶变换近红外光谱法快速测定煤炭的空干基水分[J].煤质技术,2005,6:73-74.
[56]李凤瑞,唐玉国,肖宝兰.应用近红外光谱分析技术测量煤质发热量[J].电站系统工程,2004,20(3):19-20.
[57]李凤瑞,唐玉国,肖宝兰.近红外光谱分析技术预测煤质挥发分含量模型的研究[J].热能动力工程,2003,18(6):582-583.
[58] Disla J.M.S., Janik L., McLaughlin M.J., Forrester S., Kirby J., Reimann C.. The use ofdiffuse reflectance mid-infrared spectroscopy for the prediction of the concentration ofchemical elements estimated by X-ray fluorescence in agricultural and grazing Europeansoils[J]. Applied Geochemistry,2013,29:135-143.
[59] Crabtree K.N., Hodges J.N., Siller B.M., Perry A.J., Kelly J.E., Jenkins P.A., McCall B.J..Sub-Doppler mid-infrared spectroscopy of molecular ions[J]. Chemical Physics Letters,2012,551:1-6.
[60] Capito, F., Skudas R., Kolmar H., Stanislawski B.. Host cell protein quantification byFourier transform mid-infrared spectroscopy (FT-MIR)[J]. Biotechnology&Bioengineering,2013,110(1):252-259.
[61] Ozturk B., Yucesoy D., Ozen B.. Application of mid-infrared spectroscopy for themeasurement of several quality parameters of alcoholic beverages, wine and raki[J]. FoodAnalytical Methods,2012,5(6):1435-1442.
[62] Wu Zhongchen, Tao Lin, Zhang Pengyan, Li Ping, Zhu Qianqian, Tian Yong, Du Guiqiang,Lv Maoshui, Yang Tianlin. Diffuse reflectance mid-infrared Fourier transform spectroscopyfor rapid identification of dried sea cucumber products from different geographical areas[J].Vibrational Spectroscopy,2010,53:222-226.
[63] Zhang Yu, Gao Wenzhu, Song Zhenyu, An Yupeng, Li Li, Song Zhanwei, Yu William W.,Wang Yiding. Design of a novel gas sensor structure based on mid-infrared absorptionspectrum[J]. Sensors&Actuators B,2010,147:5-9.
[64] Kos G., Lohninger H., Krska R.. Development of a method for the determination of fusariumfungi on corn using mid-infrared spectroscopy with attenuated total reflection andchemometrics[J]. Analytical Chemistry,2003,75(5):1211-1217.
[65] Geng W.H., Nakajima T., Takanashi H., Ohki A.. Analysis of carboxyl group in coal and coalaromaticity by Fourier transform infrared (FT-IR) spectrometry [J]. Fuel,2009,88:139-144.
[66] Bona M.T., Andres J.M.. Reflection and transmission mid-infrared spectroscopy for rapiddetermination of coal properties by multivariate analysis[J]. Talanta,2008,74:998-1007.
[67] Zhang Rui, Wang Wenjian. Learning linear and nonlinear PCA with linear programming[J].Neural Process Lett,2011,33:151-170.
[68]杨起,韩德馨.中国煤田地质学[M].北京:煤炭工业出版社,1982.
[69]李小明,曹代勇.不同变质类型煤的结构演化特征及其地质意义[J].中国矿业大学学报,2012,41(1):74-81.
[70]李英华.煤质分析应用技术指南(第2版)[M].北京:中国标准出版社,2009.
[71]中国国家标准化管理委员会. GB/T5751-2009我国煤炭分类[S].北京:中国标准出版社,2009.
[72]崔村丽.我国煤炭资源及其分布特征[J].科技情报开发与经济,2011,21(24):181-182.
[73]中国国家标准化管理委员会. GB475-2008商品煤样人工采取方法[S].北京:中国标准出版社,2008.
[74]中国国家标准化管理委员会. GB474-2008煤样的制备方法[S].北京:中国标准出版社,2008.
[75]中国国家标准化管理委员会. GB/T211-2007煤中全水分的测量方法[S].北京:中国标准出版社,2007.
[76]中国国家标准化管理委员会. GB/T212-2008煤的工业分析方法[S].北京:中国标准出版社,2008.
[77]中国国家标准化管理委员会. GB/T213-2008煤的发热量测定方法[S].北京:中国标准出版社,2008.
[78]刘建学.实用近红外光谱分析技术[M].北京:科学出版社,2008.
[79] Goge F., Joffre R., Jolivet C., Ross I., Ranjard L.. Optimization criteria in sample selectionstep of local regression for quantitative analysis of large soil NIRS database[J].Chemometrics&Intelligent Laboratory Systems,2012,110(1):168-176.
[80]王利,孟庆翔,任丽萍,杨建松.近红外光谱快速分析技术及其在动物饲料和产品品质检测中的应用[J].光谱学与光谱分析,2010,30(6):1482-1487.
[81] Pinto L.A., Galvao R.K.H., Araujo M.C.U.. Influence of wavelet transform settings on NIRand MIR spectrometric analyses of diesel, gasoline, corn and wheat[J]. Journal of TheBrazilian Chemical Society,2011,22(1):179-186.
[82] Prochazkova Z., Drackova M., Salakova A., Gallas L., Pospiech M., Vorlova L.,Tremlova B.,Buchtova H.. Application of FT-NIR spectroscopy in the determination of basic physical andchemical properties of sausages[J]. Acta Veterinaria Brno,2010,79(9):101-106.
[83] Corluka V., Mesic M., Valter Z.. Basic access in analysis and identification organic bonds infood industry use NIR spectrometry[J]. Actual Tasks on Agricultural Engineering,2008,36:265-273.
[84] PeanoC., Reita G., Chiabrando V.. Firmness and soluble solids assessment of nectarines byNIRs spectroscopy[C]. Proceedings of The VIth International Peach Symposium,2006,713:465-470.
[85] Nielsen J.P., Bertrand D., Micklander E., Courcoux P., Munck L.. Study of NIR spectra,particle size distributions and chemical parameters of wheat flours: a multi-way approach[J].Journal of Near Infrared Spectroscopy,2001,9(4):275-285.
[86]张小超,吴静珠,徐云.近红外光谱分析技术及其在现代农业的应用[M].北京:电子工业出版社,2012,4.
[87]郁磊.基于近红外光谱的煤质工业分析研究[D].徐州:中国矿业大学,2010,6.
[88] Liu Yande, Sun Xudong, Ouyang Aiguo. Nondestructive measurement of soluble solidcontent of navel orange fruit by visible-NIR spectrometric technique with PLSR andPCA-BPNN[J]. LWT-Food Science&Technology,2010,43:602-607.
[89] Fernandez N.J., Lopez M.I., Sanchez M.T., Morales J., Gonzalez C.V.. Shortwave-nearinfrared spectroscopy for determination of reducing sugar content during grape ripening,winemaking, and aging of white and red wines[J]. Food Research International,2009,42:285-291.
[90] Jaillais B., Pinto R., Barros A.S., Rutledge D.N.. Outer-product analysis (OPA) using PCA tostudy the influence of temperature on NIR spectra of water[J]. Vibrational Spectroscopy,2005,39(1):50-58.
[91]曾雪强.偏最小二乘降维方法的研究与应用[D].上海:上海大学,2009,12.
[92]王惠文,吴载斌,孟洁.偏最小二乘回归的线性与非线性方法[M].北京:国防工业出版社,2006,9.
[93] Leone A.P., Viscarra Rossel R.A., Amenta P., Buondonno A.. Prediction of soil propertieswith PLSR and vis-NIR spectroscopy: application to mediterranean soils from SouthernItaly[J]. Current Analytical Chemistry,2012,8(2):283-299.
[94] Blanco M., Peguero A.. A new and simple PLS calibration method for nir spectroscopy APIdetermination in intact solid formulations[J]. Analytical Methods,2012,4(6):1507-1512.
[95] Szigedi T., Lenart J., Dernovics M.. Protein content determination in Brassica oleraceaspecies using FT-NIR technique and PLS regression[J]. International Journal of FoodScience&Technology,2012,47(2):436-440.
[96] Killner M.H.M., Rohwedder J.J.R., Pasquini Celio. A PLS regression model using NIRspectroscopy for on-line monitoring of the biodiesel production reaction[J]. Fuel,2011,90(11):3268-3273.
[97] Xie Lijuan, Ye Xingqian, Liu Donghong, Ying Yibin. Quantification of glucose, fructose andsucrose in bayberry juice by NIR and PLS[J]. Food Chemistry,2009,114(3):1135-1140.
[98] Chen Quansheng, Zhao Jiewen, Liu Muhua, Cai Jianrong, Liu Jianhua. Determination oftotal polyphenols content in green tea using FT-NIR spectroscopy and different PLSalgorithms[J]. Journal of Pharmaceutical&Biomedical Analysis,2008,46(3):568-573.
[99]胡昌勤,冯艳春.近红外光谱法快速分析药品[M].北京:化学工业出版社,2010,1.
[100]蒋宗礼.人工神经网络[M].北京:高等教育出版社,2001,8.
[101] Simon H.著,申富饶,徐烨,郑俊,晁静译.神经网络与机器学习(第3版)[M].北京:机械工业出版社,2011,3.
[102]牛晓颖,邵利敏,赵志磊,张晓瑜.基于BP-ANN的草莓品种近红外光谱无损鉴别方法研究[J].光谱学与光谱分析,2012,32(8):2095-2099.
[103] Zou Qiang, Fang Hui, Liu Fei, Kong Wenwen, He Yong. Comparative study of distancediscriminant analysis and BP neural network for identification of rapeseed cultivars usingvisible/near infrared spectra[C]. Proceedings of The4th IFIP TC12Conference onComputer&Computing Technologies in Agricultur,2011,347:124-133.
[104] Qui Zhihua, Wang Lihai. Prediction of lignin content of manchurian walnut by BP neuralnetwork and near-infrared spectroscopy[C]. Proceedings of The2011InternationalConference on Materials Engineering for Advanced Technologies,2011,267:991-994.
[105]郑立华,李民赞,潘娈,孙建英,唐宁.基于近红外光谱技术的土壤参数BP神经网络预测[J].光谱学与光谱分析,2008,28(5):1160-1164.
[106]王艳斌.人工神经网络在近红外分析方法中的应用及深色油品的分析[D].北京:石油化工研究院,2002,5.
[107]李燕,王俊德,顾炳和,孟广政.人工神经网络及其在光谱分析中的应用[J].光谱学与光谱分析,1999,19(6):844-849.
[108] Yu Fan, Zhao Yingshi, Li Haitao. Soil moisture retrieval based on GA-BP neural networksalgorithm[J]. Journal of Infrared&Millimeter Waves,2012,31(3):283-288.
[109] Shifei Ding, Chunyang Su, Junzhao Yu. An optimizing BP neural network algorithm basedon genetic algorithm[J]. Artificial Intelligence Review,2011,36(2):153-162.
[110]雷萌,李明,徐志彬.遗传神经网络在近红外光谱煤质分析中的应用研究[J].工矿自动化,2010,2:41-43.
[111]雷萌,李明,吴楠,李颖娜,程玉虎.粒度对煤质近红外定量分析模型的影响[J].光谱学与光谱分析,2013,33(1):65-68.
[112] Yang Haiqing, Kuang Boyan, Mouazen A.M.. Selection of preprocessing parameters forPCA of soil classification affected by particle sizes based on vis/NIR spectroscopy[J]. KeyEngineering Materials,2011,467-469:725-730.
[113] Muller S.B., Zwicky C.N., Blahous L., Near-infrared (NIR) on-line analysis for coarse-grained raw materials[J]. ZKG International,2011,64(4):44-53.
[114] Stenberg B. Effects of soil sample pretreatments and standardised rewetting as interactedwith sand classes on vis-NIR predictions of clay and soil organic carbon[J]. Geoderma,2010,158(1-2):15-22.
[115] Blanco M., Peguero A.. Influence of physical factors on the accuracy of calibration modelsfor NIR spectroscopy[J]. Journal of Pharmaceutical&Biomedical Analysis,2010,52(1):59-65.
[116] Wang Dong, Ma Zhihong, He Hongju, Pan Ligang, Min Shungeng, Wang Jihua. Study onthe influence of outlier elimination in the quantitative near-infrared models of paprikaquality[J]. Sensor Letters,2012,10(1-2):162-167.
[117]王建义,雷萌.近红外光谱煤质分析模型中异常样品的剔除方法[J].工矿自动化,2011,1:75-77.
[118] Liu Zhichao, Cai Wensheng, Shao Xueguang. Outlier detection in near-infraredspectroscopic analysis by using Monte Carlo cross-validation[J]. Science in China SeriesB-Chemistry,2008,51(8):751-759.
[119]刘强,罗长兵,陈绍江,孟庆翔.近红外光谱分析青贮玉米NDF中判别异常光谱的研究[J].2007,27(8):1514-1518.
[120] Seungdo J. Sangwook K., Byunguk C.. Dimensionality reduction for similarity search withthe Euclidean distance in high-dimensional applications[J].Multimedia Tools&Applications,2009,42(2):251-271.
[121] Mahalanobis P.C.. On the generalized distance in statistics[C]. Proceedings of The NationalInstitute of Science of India(NISI),1936,12:49-55.
[122]周大镯.多变量时间序列的聚类、相似查询与异常检测[D].天津:天津大学,2008,12.
[123] Hawkins D..Identification of outliers[M]. London: Chapman&Hall,1980.
[124] Ye Jun. Cosine similarity measures for intuitionistic fuzzy sets and their applications[J].Mathematical&Computer Modeling,2011,53(1-2):91-97.
[125] Zhu Shiwei, Wu Junjie, Xiong Hui, Xia Guoping. Scaling up top-K cosine similaritysearch[J]. Data&Knowledge Engineering,2011,70:60-83.
[126] Liu Chengjun. Clarification of assumptions in the relationship between the Bayes decisionrule and the whitened cosine similarity measure[J]. IEEE Transactions on Pattern Analysis&Machine Intelligence,2008,30(6):1116-1117.
[127] Sergios T., Konstantinos K.著,李晶皎,王爱侠,张广渊等译.模式识别(第4版)[M].北京:电子工业出版社,2010,2.
[128]王钰,周志华,周傲英.机器学习及其应用[M].北京:清华大学出版社,2006,3.
[129] Cover T.M., Hart P.E.. Nearest neighbor pattern classification[J]. IEEE Transaction onInformation Theory,1967,6:113-121.
[130]陆婉珍.现代近红外光谱分析技术(第2版)[M].北京:中国石化出版社,2007,1.
[131]同济大学数学系.高等数学(第6版,上册)[M].北京:高等教育出版社,2007.
[132] Geladi P., Macdougall D., Martens H.. Linearization and scatter-correction for near-infraredreflectance spectra of meat[J]. Applied Spectroscopy,1985,39(3):491-500.
[133] Isaksson T., Kowalski B.R.. Piece-wise multiplicative scatter correction applied tonear-infrared transmittance data from meat products[J]. Applied Spectroscopy,1993,47(6):702-709.
[134] Windig W., Shaver J., Bro R.. Loopy MSC: a simple way to improve multiplicative scattercorrection[J]. Applied Spectroscopy,2008,62(10):1153-1159.
[135]芦永军,曲艳玲,宋敏.近红外相关光谱的多元散射校正处理研究[J].光谱学与光谱分析,2007,27(5):877-880.
[136] Maleki M.R., Mouazen A.M., Ramon H., Baerdemaeker J.D.. Multiplicative scattercorrection during on-line measurement with near infrared spectroscopy[J]. BiosystemsEngineering,2007,96(3):427-433.
[137] Thennadil S.N., Martens H., Kohler A.. Physics-based multiplicative scatter correctionapproaches for improving the performance of calibration models[J]. Applied Spectroscopy,2006,60(3):315-321.
[138] Chen J.Y., Iyo C.. Effect of multiplicative scatter correction on wavelength selection for nearinfrared calibration to determine fat content in raw milk[J]. Journal of Near InfraredSpectroscopy,2002,10(4):301-307.
[139]靳刘蕊.函数性数据分析方法及应用研究[D].厦门:厦门大学,2008,3.
[140]梁逸曾,俞汝勤.分析化学手册(10)——化学计量学[M].北京:化工出版社,2001.
[141] Tao Pan, Jiang Guoqiang, Chen Jiemei. Waveband Selection of NIR Spectroscopy Analysisfor Glucose Aqueous Solution Based on Savitzky-Golay Smoothing[J]. Advanced MaterialsResearch,2011,181-182:712-716.
[142] Delwiche S.R., Reeves J.B.. A graphical method to evaluate spectral preprocessing inmultivariate regression calibrations: Example with Savitzky-Golay filters and partial leastsquares regression[J]. Applied Spectroscopy,2010,64(1):73-82.
[143]谢军,潘涛,陈洁梅,陈华舟,任小焕.血糖近红外光谱分析的Savitzky-Golay平滑模式与偏最小二乘法因子数的联合优选[J].分析化学,2010,38(3):342-346.
[144]褚小立,袁洪福,陆婉珍.近红外分析中光谱预处理及波长选择方法进展与应用[J].化学进展,2004,16(4):528-541.
[145] Savitzky A., Golay M.J.E.. Smoothing and differentiation of data by simplified least squaresprocedures[J]. Analytical Chemometrics,1964,36(8):1627-1639.
[146] Gorry P.A.. General least-squares smoothing and differentiation by the convolution(Savitzky-Golay) method [J]. Analytical Chemometrics,1990,62:570-573.
[147] Huang L.H.. A roughness-penalty view of kernel smoothing[J]. Statistics&ProbabilityLetters,2001,52(1):85-89.
[148] Kennedy J., Eberhart R.C.. Particle swarm optimization[C]. Proceedings of The IEEEInternational Conference on Neural Networks,1995,4:1942-1948.
[149] Shi Y., Eberhart R.C.. A modified particle swarm optimizer[C]. Proceedings of The IEEEInternational Conference on Evolutionary Computation,1998:69-73.
[150] Eberhart R.C., Shi Y.H.. Particle swarm optimization: Developments, applications andresources[C]. Proceedings of The2001Congress on Evolutionary Computation,2001,1-2:81-86.
[151] Ethem A.著,范明译.机器学习导论[M].北京:机械工业出版社,2009,6.
[152] Blanchet F.G., Legendre P., Borcard D.. Forward selection of explanatory variables[J].Ecology,2008,89(9):2623-2632.
[153] Whitley D.C., Ford M.G., Livingstone D.J.. Unsupervised forward selection: A method foreliminating redundant variables[J]. Journal of Chemical Information&Computer Sciences,2000,40(5):1160-1168.
[154]徐云.农产品品质检测中的近红外光谱分析方法研究[D].北京:中国农业大学,2009.
[155] Holland J. H.. Adaptation in nature and artificial systems[M]. Michigan: The University ofMichigan Press,1975.
[156] Bagley J. D.. The behavior of adaptive system which employs genetic and correlationalgorithm[M]. Michigan: The University of Michigan Press,1967.
[157] Goldberg D.. Genetic algorithm in search, optimization and machine learning[M]. Boston:Addison Wesley Longman Publishing Co.,1989.
[158] Davis L.. Handbook of genetic algorithms[M]. New York: Van Nostrand Reinhold,1991.
[159]张顶学.遗传算法与粒子群算法的改进及应用[D].武汉:华中科技大学,2007.
[160] Fei Qiang, Li Ming, Wang Bin, Huan Yanfu, Feng Guodong, Ren Yulin. Analysis ofcefalexin with NIR spectrometry coupled to artificial neural networks with modified geneticalgorithm for wavelength selection[J]. Chemometrics&Intelligent Laboratory Systems,2009,97(2):127-131.
[161] Xu Zhibin, Yu Lei, Lei Meng, An Baoran. Application research on coal analysis of nearinfrared spectroscopy (NIRS) by intelligent algorithms[C]. Proceedings of The22nd ChineseControl&Decision Conference,2010,2416-2419.
[162] Shi Bolin, Ji Baoping, Zhu Dazhou, Tu Zhenhua, Qing Zhaoshen. Study on geneticalgorithms-based nir wavelength selection for determination of soluble solids content in fujiapples[J]. Journal of Food Quality,2008,31(2):232-249.
[163] Reynes Ch., de Souza S., Sabatier R., Figueres G., Vidal B.. Selection of discriminantwavelength intervals in NIR spectrometry with genetic algorithms[J]. Journal ofChemometrics,2006,20(3-4):136-145.
[164] Wu Jianning, Wang Jue, Liu Li. Feature extraction via KPCA for classification of gaitpatterns[J]. Human movement science,2007,26:393-411.
[165] Michael B.R., Indra A.. Classification and regionalization through kernel principalcomponent analysis[J]. Physics&Chemistry of The Earth,2010,35:316-328.
[166] Fu G., Frank Y. Shih, Wang H.M.. A kernel-based parametric method for conditional densityestimation[J]. Pattern Recognition,2011,44:284-294.
[167] Zhao Lihong, Zhang Xili, Xu Xinhe. Face recognition base on KPCA with polynomialkernels[C]. Proceedings of The5th International Conference on Wavelet Analysis&PatternRecognition,2007,1-4:1213-1216.
[168] Jorgensen K.W., Hansen L.K.. Model selection for gaussian kernel PCA denoising[J]. IEEETransactions on Neural Networks&Learning Systems,2012,23(1):163-168.
[169] Kohonen, T.. Essentials of the self-organizing map[J]. Neural Networks,2013,37:52-65.
[170] Vesanto J., Alhoniemi E.. Clustering of the self-organizing map[J]. IEEE Transactions onNeural Networks,2000,11(3):586-600.
[171] Creput J.C., Hajjam A., Koukam A., Kuhn O.. Self-organizing maps in population basedmetaheuristic to the dynamic vehicle routing problem[J]. Journal of CombinatorialOptimization,2012,24(4):437-458.
[172] Shieh S., Liao I.. A new approach for data clustering and visualization using self-organizingmaps[J]. Expert Systems with Applications,2012.39(15):11924-11933.
[173] Ballabio D., Vasighi M.. A MATLAB toolbox for self organizing maps and supervised neuralnetwork learning strategies[J]. Chemometrics&Intelligent Laboratory Systems,2012,118:24-32.
[174] Hong Yingyi, Chou Jinghan. Nonintrusive energy monitoring for microgrids using hybridself-organizing feature-mapping networks[J]. Energies,2012,5(7):2578-2593.
[175] Huang G.B., Saratchandran P., Sundararajan N.. A generalized growing and pruning RBF(GGAP-RBF) neural network for function approximation[J]. IEEE Transactions on NeuralNetworks,16(1):57-67.
[176] Telmoudi A.J., Tlijani H., Nabli L., Ali M., M'Hiri R.. A new RBF neural network forprediction in industrial control[J]. International Journal of Information Technology&Decision Making,2012,11(4):749-775.
[177] She Yuanguo. Hybrid intelligent algorithm of RBF neural network with variable lengthstructure[J]. Information-An International Interdisciplinary Journal,2012,15(6):2571-2577.
[178] Han Honggui, Qiao Junfei. Adaptive computation algorithm for RBF neural network[J].IEEE Transactions on Neural Networks&Learning Systems,2012,23(2):342-347.
[179] Du Dajun, Li Kang, Fei Minrui. A fast multi-output RBF neural network constructionmethod[J]. Neurocomputing,2010,73(10-12):2196-2202.
[180] Lin C.L., Wang J.F., Chen C.Y., Chen C.W., Yen C.W.. Improving the generalizationperformance of RBF neural networks using a linear regression technique[J]. Expert Systemswith Applications,2009,36(10):12049-12053.
[181] Georgia M., Antreas A., Kalliopi M., Haralambos S., Olga I.M.. Prediction of toxicity usinga novel RBF neural network training methodology[J]. Journal of Molecular Modeling,2006,12(3):297-305.
[182] Dietterich T.G.. Machine Leaning Research: Four Current Directions[J]. AI Magazine,1997,18(4):97-136.
[183] Hansen L.K., Salamon P.. Neural Network Ensembles[J]. IEEE Transactions on PatternAnalysis&Machine Intelligence,1990,12(10):993-1001.
[184] Ahn B.S., Cho S.S., Kim C.Y.. The integrated methodology of rough set theory and artificialneural network for business failure prediction[J]. Expert Systems with Applications,2000,18(2):65-74.
[185] Chen J.. A predictive system for blast furnaces by integrating a neural network withqualitative analysis[J]. Engineering Applications of Artificial Intelligence,2001,14(1):77-85.
[186]沈掌泉.神经网络集成技术及其在土壤学中应用的研究[D].杭州:浙江大学,2005,5.
[187] Dor O., Zhou Y.Q.. Real-SPINE: An integrated system of neural networks for real-valueprediction of protein structural properties[J]. Proteins-Structure Function&Bioinformatics,2007,68(1):76-81.
[188] Xue B., Faraggi E., Zhou Y.. Predicting residue-residue contact maps by a two-layer,integrated neural-network method[J]. Proteins-Structure Function&Bioinformatics,2009,76(1):176-183.
[189] Azadeh A., Maghsoudi A., Sohrabkhani S.. An integrated artificial neural networks approachfor predicting global radiation[J]. Energy Conversion&Management,2009,50(6):1497-1505.
[2]何选明.煤化学(第2版)[M].北京:冶金工业出版社,2010,5.
[3]邢志良.煤炭生产中全过程质量控制方法研究[J].中国矿业,2010,19(3):44-47.
[4]滕吉文,张雪梅,杨辉.中国主体能源——煤炭的第二深度空间勘探、开发和高效利用[J].地球物理学进展,2008,23(4):972-992.
[5]陈建国,张庆福,马玖坤,张鹤鹏.加强煤炭质量管理提高商品煤销售效益[J].煤炭技术,2005,24(2):119-120.
[6]张双全,吴国光.煤化学[M].徐州:中国矿业大学出版社,2004.
[7]于实,李丰田.煤质检测分析新技术新方法与化验结果的审查计算实用手册[M].北京:当代中国音像出版社,2005,5.
[8]施玉英.煤炭分析试验仪器设备的使用与维修[M].北京:煤炭工业出版社,2007,8.
[9]陆婉珍,袁洪福,楮小立.近红外光谱仪器[M].北京:化学工业出版社,2010,2.
[10] Yasukuni R., Boubekri R., Grand J., Felidj N., Maurel F., Perrier A., Metivier R., NakataniK., Yu P., Aubard J.. Specific and nondestructive detection of different diarylethene isomersby NIR-SERS[J]. Journal of Physical Chemistry C,2012,116(30):16063-16069.
[11] Bahareh J., Saeid M., Ezzedin M., Hassan G.. Reflectance Vis/NIR spectroscopy fornondestructive taste characterization of Valencia oranges[J]. Computers&Electronics inAgriculture,2012,85:64-69.
[12] Ni Yong, Shao Yongni, He Yong. Fast nondestructive measurement and discrimination oftablet based on Vis/NIR spectroscopy and chemometrics[C]. Proceedings of The3rdInternational Conference on Measuring Technology&Mechatronics Automation,2011,48-49:506-510.
[13] Raúl G., Antonio J.S., J.G., Eugenio I., Ana F., Jose M.B.. Nondestructive assessment offreshness in packaged sliced chicken breasts using SW-NIR spectroscopy[J]. Food ResearchInternational,2010,44(1):331-337.
[14] Tripathi S., Patel K.G., Bafna A.M.. Nondestructive determination of curcuminoids fromturmeric powder using FT-NIR[J]. Journal of Food Science&Technology-Mysore,2010,47(6):678-681
[15]谢晓霞,付利俊,庞文娟.影响煤工业分析特性指标检测的关键步骤[J].煤质技术,2010,6:33-34.
[16] Gu Deshan, Cheng Daowen, Yin Jie, Qiao Shuang, Liu Linmao, Zhang Xuesong. Rapidanalysis of coal properties[J]. IEEE Instrumentation&Measurement Magazine,2005,12:22-27.
[17] Xia Wencheng, Yang Jianguo, Liang Chuan. Effect of microwave pretreatment on oxidizedcoal flotation[J]. Powder Technology,2013,233:186-189
[18] Fu Huilin, Shen Zhongli. Microwave heating-based analysis method for the coal moisturecontent[J]. Energy Education Science&Technology Part A-energy Science&Research,2011,28(1):293-300.
[19] Arash T., Yu J.L., Li X.Ch., Meesri C.. Experimental study on microwave drying of Chineseand Indonesian low-rank coals[J]. Fuel Processing Technology,2011,92(10):1821-1829.
[20] Ozbayoglu G., Depci T., Ataman N.. Effect of microwave radiation on coal flotation[J].Energy Sources Part A-Recovery Utilization&Environmental Effects,2009,31(6):492-499.
[21] Liu Haiyu, Tan Houzhang, Gao Qiang, Hu Hongli, Xu Tongmo. Experimental study of fastmeasuring moisture content of coal by microwave[C]. Proceedings of The7th InternationalSymposium on Test&Measurement,2007,1-7:1937-1941.
[22]王辉东.中子活化煤质在线分析及β辐射伏特核电池能量输运研究[D].吉林大学,2012.
[23] He L.L., Melnichenko Y.B., Mastalerz M., Sakurovs R., Radlinski A.P., Blach Tomas. PoreAccessibility by methane and carbon dioxide in coal as determined by neutron scattering[J].Energy&Fuels,2012,26(3):1975-1983.
[24] Rey R.M.A., Alonso S.T., Castro G.M.P.. Neutron activation of deferred gamma rays steps todetermine the usefulness of this technique in analyzing elements of a mineral sample[C].Proceedings of The5th Meeting of the Spanish Neutron Scattering Association,2011,325:12025-12029.
[25] Lim C.S., Abernethy D.A.. On-line coal analysis using fast neutron-induced gamma-rays[J].Applied Radiation&Isotopes,2005,63(5-6):697-704.
[26] Akkaya, A.V.. Predicting coal heating values using proximate analysis via a neural networkapproach[J]. Energy Sources Part A-Recovery Utilization&Environmental Effects,2013,35(3):253-260.
[27]王建军,王世营,雷萌.粒子群优化算法在煤炭发热量预测模型中的应用[J].工矿自动化,2012,5:50-53.
[28] Chelgani S.C., Hart B., Grady W.C., Hower J.C.. Study relationship between inorganic andorganic coal analysis with gross calorific value by multiple regression and ANFIS[J].International Journal of Coal Preparation&Utilization,2011,31(1):9-19.
[29] Yao Wanye, Su Ling, Yin Shi. The application of BP neural network in coal analysis[C].Proceedings of The2011International Conference on Machine Learning&Cybernetics,2011,7:795-798.
[30] Jorjani E., Poorali H.A., Sam A., Chelgani S. Ch., Mesroghli Sh., Shayestehfar M.R..Prediction of coal response to froth flotation based on coal analysis using regression andartificial neural network[J]. Minerals Engineering,2009,22(11):970-976.
[31]周翠红,路迈西.线性回归与人工神经网络预测煤炭发热量[J].煤炭科学技术,2009,37(2):117-120.
[32]江文豪,韦红旗,屈天章,朱锋.基于遗传算法优化参数的支持向量机燃煤发热量预测[J].热力发电,2011,40(3):14-19.
[33] Burgess C., Hammond J.. Exploring the "forgotten" region-an update on NIRspectrometry[J]. Spectroscopy Europe,2009,21(2):28-31.
[34]严衍禄,赵龙莲,韩东海,杨曙明.近红外光谱分析基础及应用[M].北京:中国轻工业出版社,2005,1.
[35] Blanco M., Villarroya I.. NIR spectroscopy: a rapid-response analytical tool[J]. Trends inAnalytical Chemistry,2002,21(4):240-250.
[36]徐广通,袁洪福,陆婉珍.近红外光谱定量校正模型适用性研究[J].光谱学与光谱分析2001,21(4):459-463.
[37]徐广通,袁洪福,陆婉珍.现代近红外光谱技术及应用进展[J].光谱学与光谱分析,2000,20(2):134-142.
[38] Salgo A., Gergely S.. Analysis of wheat grain development using NIR spectroscopy[J].Journal of Cereal Science,2011,56(1):31-38.
[39] Shetty N., Gislum R., Jensen A.M.D., Boelt B.. Development of NIR calibration models toassess year-to-year variation in total non-structural carbohydrates in grasses using PLSR[J].Chemometrics&Intelligent Laboratory Systems,2012,111(1):34-38.
[40] Lin Hao, Zhao Jiewen, Sun Li, Chen Quansheng, Zhou Fang. Freshness measurement ofeggs using near infrared spectroscopy and multivariate data analysis[J]. Innovative FoodScience&Emerging Technologies,2011,12(2):182-186.
[41] Lourdes S.Ch., Vincent B., Ouissam A., Francisco P.R.. On-line analysis of intact olive fruitsby vis-NIR spectroscopy: Optimisation of the acquisition parameters[J]. Journal of FoodEngineering,2012,112(3):152-157.
[42] Kurihara K., Kawaguchi H., Obata T., Ito H., Sakatani K., Okada E.. The influence of frontalsinus in brain activation measurements by near-infrared spectroscopy analyzed by realistichead models[J]. Biomedical Optics Express,2012,3(9):2121-2130.
[43] De Bleye C., Chavez P.F., Mantanus J., Marini R., Hubert Ph., Rozet E., Ziemons E.. Criticalreview of near-infrared spectroscopic methods validations in pharmaceutical applications[J].Journal of Pharmaceutical&Biomedical Analysis,2012,69(SI):125-132.
[44] Prieto N., Dugan M.E.R., Lopez C.O., McAllister, T.A., Aalhus, J.L., Uttaro, B.. Nearinfrared reflectance spectroscopy predicts the content of polyunsaturated fatty acids andbiohydrogenation products in the subcutaneous fat of beef cows fed flaxseed[J]. MeatScience,2012,90(1):43-51.
[45] Kodzue K., Hiroyuki M., Mari M., Natsuki H., Hideyasu K., Hiroshi K., Wang P.Y., OsamuI., Hiroshi K., Roumiana T.. Near infrared spectroscopy of urine proves useful for estimatingovulation in giant panda (ailuropoda melanoleuca)[J]. Analytical Methods,2010,2:1671-1675.
[46]孔翠萍,褚小立,杜泽学,陆婉珍.近红外光谱方法预测生物柴油主要成分[J].分析化学,2010,38(6):805-810.
[47] Kaihara M., Takahashi T., Akazawa T., Sato T., Takahashi S.. Application of near infraredspectroscopy to rapid analysis of coals[J]. Spectroscopy Letters,2002,35(3):369-376.
[48] Andres J.M., Bona M.T.. Analysis of coal by diffuse reflectance near-infrared spectroscopy[J]. Analytica Chimica Acta,2005,535:123-132.
[49] Andres J.M., Bona M.T.. Coal analysis by diffuse reflectance near-infrared spectroscopy:Hierarchical cluster and linear discrimination analysis[J]. Talanta,2007,72:1423-1431.
[50] Andres J.M., Bona M.T.. ASTM clustering for improving coal analysis by near-infraredspectroscopy[J]. Talanta,2006,70:711-719.
[51] Dong Won Kim, Jong Min Lee, Jae Sung Kim. Application of near infrared diffusereflectance spectroscopy for on-line measurement of coal properties[J]. Korean Journal ofChemical Engineering,2009,26(2):489-495.
[52]武中臣,熊智新,王海东,张淑宁,张鹏彦,贾滨,贺守波.褐煤品质的傅里叶变换近红外光谱定量分析[J].光谱实验室,2008,25(5):819-823.
[53]丁仁杰,张笑.基于近红外光谱分析的电厂煤质分类快速检测方法[J].电站系统工程,2007,23(1):32-34.
[54]邬蓓蕾,林振兴,王群威,林力.傅立叶变换近红外光谱定量分析煤炭挥发分[J].岩矿测试,2006,25(2):133-136.
[55]邬蓓蕾,林振兴,王群威.傅立叶变换近红外光谱法快速测定煤炭的空干基水分[J].煤质技术,2005,6:73-74.
[56]李凤瑞,唐玉国,肖宝兰.应用近红外光谱分析技术测量煤质发热量[J].电站系统工程,2004,20(3):19-20.
[57]李凤瑞,唐玉国,肖宝兰.近红外光谱分析技术预测煤质挥发分含量模型的研究[J].热能动力工程,2003,18(6):582-583.
[58] Disla J.M.S., Janik L., McLaughlin M.J., Forrester S., Kirby J., Reimann C.. The use ofdiffuse reflectance mid-infrared spectroscopy for the prediction of the concentration ofchemical elements estimated by X-ray fluorescence in agricultural and grazing Europeansoils[J]. Applied Geochemistry,2013,29:135-143.
[59] Crabtree K.N., Hodges J.N., Siller B.M., Perry A.J., Kelly J.E., Jenkins P.A., McCall B.J..Sub-Doppler mid-infrared spectroscopy of molecular ions[J]. Chemical Physics Letters,2012,551:1-6.
[60] Capito, F., Skudas R., Kolmar H., Stanislawski B.. Host cell protein quantification byFourier transform mid-infrared spectroscopy (FT-MIR)[J]. Biotechnology&Bioengineering,2013,110(1):252-259.
[61] Ozturk B., Yucesoy D., Ozen B.. Application of mid-infrared spectroscopy for themeasurement of several quality parameters of alcoholic beverages, wine and raki[J]. FoodAnalytical Methods,2012,5(6):1435-1442.
[62] Wu Zhongchen, Tao Lin, Zhang Pengyan, Li Ping, Zhu Qianqian, Tian Yong, Du Guiqiang,Lv Maoshui, Yang Tianlin. Diffuse reflectance mid-infrared Fourier transform spectroscopyfor rapid identification of dried sea cucumber products from different geographical areas[J].Vibrational Spectroscopy,2010,53:222-226.
[63] Zhang Yu, Gao Wenzhu, Song Zhenyu, An Yupeng, Li Li, Song Zhanwei, Yu William W.,Wang Yiding. Design of a novel gas sensor structure based on mid-infrared absorptionspectrum[J]. Sensors&Actuators B,2010,147:5-9.
[64] Kos G., Lohninger H., Krska R.. Development of a method for the determination of fusariumfungi on corn using mid-infrared spectroscopy with attenuated total reflection andchemometrics[J]. Analytical Chemistry,2003,75(5):1211-1217.
[65] Geng W.H., Nakajima T., Takanashi H., Ohki A.. Analysis of carboxyl group in coal and coalaromaticity by Fourier transform infrared (FT-IR) spectrometry [J]. Fuel,2009,88:139-144.
[66] Bona M.T., Andres J.M.. Reflection and transmission mid-infrared spectroscopy for rapiddetermination of coal properties by multivariate analysis[J]. Talanta,2008,74:998-1007.
[67] Zhang Rui, Wang Wenjian. Learning linear and nonlinear PCA with linear programming[J].Neural Process Lett,2011,33:151-170.
[68]杨起,韩德馨.中国煤田地质学[M].北京:煤炭工业出版社,1982.
[69]李小明,曹代勇.不同变质类型煤的结构演化特征及其地质意义[J].中国矿业大学学报,2012,41(1):74-81.
[70]李英华.煤质分析应用技术指南(第2版)[M].北京:中国标准出版社,2009.
[71]中国国家标准化管理委员会. GB/T5751-2009我国煤炭分类[S].北京:中国标准出版社,2009.
[72]崔村丽.我国煤炭资源及其分布特征[J].科技情报开发与经济,2011,21(24):181-182.
[73]中国国家标准化管理委员会. GB475-2008商品煤样人工采取方法[S].北京:中国标准出版社,2008.
[74]中国国家标准化管理委员会. GB474-2008煤样的制备方法[S].北京:中国标准出版社,2008.
[75]中国国家标准化管理委员会. GB/T211-2007煤中全水分的测量方法[S].北京:中国标准出版社,2007.
[76]中国国家标准化管理委员会. GB/T212-2008煤的工业分析方法[S].北京:中国标准出版社,2008.
[77]中国国家标准化管理委员会. GB/T213-2008煤的发热量测定方法[S].北京:中国标准出版社,2008.
[78]刘建学.实用近红外光谱分析技术[M].北京:科学出版社,2008.
[79] Goge F., Joffre R., Jolivet C., Ross I., Ranjard L.. Optimization criteria in sample selectionstep of local regression for quantitative analysis of large soil NIRS database[J].Chemometrics&Intelligent Laboratory Systems,2012,110(1):168-176.
[80]王利,孟庆翔,任丽萍,杨建松.近红外光谱快速分析技术及其在动物饲料和产品品质检测中的应用[J].光谱学与光谱分析,2010,30(6):1482-1487.
[81] Pinto L.A., Galvao R.K.H., Araujo M.C.U.. Influence of wavelet transform settings on NIRand MIR spectrometric analyses of diesel, gasoline, corn and wheat[J]. Journal of TheBrazilian Chemical Society,2011,22(1):179-186.
[82] Prochazkova Z., Drackova M., Salakova A., Gallas L., Pospiech M., Vorlova L.,Tremlova B.,Buchtova H.. Application of FT-NIR spectroscopy in the determination of basic physical andchemical properties of sausages[J]. Acta Veterinaria Brno,2010,79(9):101-106.
[83] Corluka V., Mesic M., Valter Z.. Basic access in analysis and identification organic bonds infood industry use NIR spectrometry[J]. Actual Tasks on Agricultural Engineering,2008,36:265-273.
[84] PeanoC., Reita G., Chiabrando V.. Firmness and soluble solids assessment of nectarines byNIRs spectroscopy[C]. Proceedings of The VIth International Peach Symposium,2006,713:465-470.
[85] Nielsen J.P., Bertrand D., Micklander E., Courcoux P., Munck L.. Study of NIR spectra,particle size distributions and chemical parameters of wheat flours: a multi-way approach[J].Journal of Near Infrared Spectroscopy,2001,9(4):275-285.
[86]张小超,吴静珠,徐云.近红外光谱分析技术及其在现代农业的应用[M].北京:电子工业出版社,2012,4.
[87]郁磊.基于近红外光谱的煤质工业分析研究[D].徐州:中国矿业大学,2010,6.
[88] Liu Yande, Sun Xudong, Ouyang Aiguo. Nondestructive measurement of soluble solidcontent of navel orange fruit by visible-NIR spectrometric technique with PLSR andPCA-BPNN[J]. LWT-Food Science&Technology,2010,43:602-607.
[89] Fernandez N.J., Lopez M.I., Sanchez M.T., Morales J., Gonzalez C.V.. Shortwave-nearinfrared spectroscopy for determination of reducing sugar content during grape ripening,winemaking, and aging of white and red wines[J]. Food Research International,2009,42:285-291.
[90] Jaillais B., Pinto R., Barros A.S., Rutledge D.N.. Outer-product analysis (OPA) using PCA tostudy the influence of temperature on NIR spectra of water[J]. Vibrational Spectroscopy,2005,39(1):50-58.
[91]曾雪强.偏最小二乘降维方法的研究与应用[D].上海:上海大学,2009,12.
[92]王惠文,吴载斌,孟洁.偏最小二乘回归的线性与非线性方法[M].北京:国防工业出版社,2006,9.
[93] Leone A.P., Viscarra Rossel R.A., Amenta P., Buondonno A.. Prediction of soil propertieswith PLSR and vis-NIR spectroscopy: application to mediterranean soils from SouthernItaly[J]. Current Analytical Chemistry,2012,8(2):283-299.
[94] Blanco M., Peguero A.. A new and simple PLS calibration method for nir spectroscopy APIdetermination in intact solid formulations[J]. Analytical Methods,2012,4(6):1507-1512.
[95] Szigedi T., Lenart J., Dernovics M.. Protein content determination in Brassica oleraceaspecies using FT-NIR technique and PLS regression[J]. International Journal of FoodScience&Technology,2012,47(2):436-440.
[96] Killner M.H.M., Rohwedder J.J.R., Pasquini Celio. A PLS regression model using NIRspectroscopy for on-line monitoring of the biodiesel production reaction[J]. Fuel,2011,90(11):3268-3273.
[97] Xie Lijuan, Ye Xingqian, Liu Donghong, Ying Yibin. Quantification of glucose, fructose andsucrose in bayberry juice by NIR and PLS[J]. Food Chemistry,2009,114(3):1135-1140.
[98] Chen Quansheng, Zhao Jiewen, Liu Muhua, Cai Jianrong, Liu Jianhua. Determination oftotal polyphenols content in green tea using FT-NIR spectroscopy and different PLSalgorithms[J]. Journal of Pharmaceutical&Biomedical Analysis,2008,46(3):568-573.
[99]胡昌勤,冯艳春.近红外光谱法快速分析药品[M].北京:化学工业出版社,2010,1.
[100]蒋宗礼.人工神经网络[M].北京:高等教育出版社,2001,8.
[101] Simon H.著,申富饶,徐烨,郑俊,晁静译.神经网络与机器学习(第3版)[M].北京:机械工业出版社,2011,3.
[102]牛晓颖,邵利敏,赵志磊,张晓瑜.基于BP-ANN的草莓品种近红外光谱无损鉴别方法研究[J].光谱学与光谱分析,2012,32(8):2095-2099.
[103] Zou Qiang, Fang Hui, Liu Fei, Kong Wenwen, He Yong. Comparative study of distancediscriminant analysis and BP neural network for identification of rapeseed cultivars usingvisible/near infrared spectra[C]. Proceedings of The4th IFIP TC12Conference onComputer&Computing Technologies in Agricultur,2011,347:124-133.
[104] Qui Zhihua, Wang Lihai. Prediction of lignin content of manchurian walnut by BP neuralnetwork and near-infrared spectroscopy[C]. Proceedings of The2011InternationalConference on Materials Engineering for Advanced Technologies,2011,267:991-994.
[105]郑立华,李民赞,潘娈,孙建英,唐宁.基于近红外光谱技术的土壤参数BP神经网络预测[J].光谱学与光谱分析,2008,28(5):1160-1164.
[106]王艳斌.人工神经网络在近红外分析方法中的应用及深色油品的分析[D].北京:石油化工研究院,2002,5.
[107]李燕,王俊德,顾炳和,孟广政.人工神经网络及其在光谱分析中的应用[J].光谱学与光谱分析,1999,19(6):844-849.
[108] Yu Fan, Zhao Yingshi, Li Haitao. Soil moisture retrieval based on GA-BP neural networksalgorithm[J]. Journal of Infrared&Millimeter Waves,2012,31(3):283-288.
[109] Shifei Ding, Chunyang Su, Junzhao Yu. An optimizing BP neural network algorithm basedon genetic algorithm[J]. Artificial Intelligence Review,2011,36(2):153-162.
[110]雷萌,李明,徐志彬.遗传神经网络在近红外光谱煤质分析中的应用研究[J].工矿自动化,2010,2:41-43.
[111]雷萌,李明,吴楠,李颖娜,程玉虎.粒度对煤质近红外定量分析模型的影响[J].光谱学与光谱分析,2013,33(1):65-68.
[112] Yang Haiqing, Kuang Boyan, Mouazen A.M.. Selection of preprocessing parameters forPCA of soil classification affected by particle sizes based on vis/NIR spectroscopy[J]. KeyEngineering Materials,2011,467-469:725-730.
[113] Muller S.B., Zwicky C.N., Blahous L., Near-infrared (NIR) on-line analysis for coarse-grained raw materials[J]. ZKG International,2011,64(4):44-53.
[114] Stenberg B. Effects of soil sample pretreatments and standardised rewetting as interactedwith sand classes on vis-NIR predictions of clay and soil organic carbon[J]. Geoderma,2010,158(1-2):15-22.
[115] Blanco M., Peguero A.. Influence of physical factors on the accuracy of calibration modelsfor NIR spectroscopy[J]. Journal of Pharmaceutical&Biomedical Analysis,2010,52(1):59-65.
[116] Wang Dong, Ma Zhihong, He Hongju, Pan Ligang, Min Shungeng, Wang Jihua. Study onthe influence of outlier elimination in the quantitative near-infrared models of paprikaquality[J]. Sensor Letters,2012,10(1-2):162-167.
[117]王建义,雷萌.近红外光谱煤质分析模型中异常样品的剔除方法[J].工矿自动化,2011,1:75-77.
[118] Liu Zhichao, Cai Wensheng, Shao Xueguang. Outlier detection in near-infraredspectroscopic analysis by using Monte Carlo cross-validation[J]. Science in China SeriesB-Chemistry,2008,51(8):751-759.
[119]刘强,罗长兵,陈绍江,孟庆翔.近红外光谱分析青贮玉米NDF中判别异常光谱的研究[J].2007,27(8):1514-1518.
[120] Seungdo J. Sangwook K., Byunguk C.. Dimensionality reduction for similarity search withthe Euclidean distance in high-dimensional applications[J].Multimedia Tools&Applications,2009,42(2):251-271.
[121] Mahalanobis P.C.. On the generalized distance in statistics[C]. Proceedings of The NationalInstitute of Science of India(NISI),1936,12:49-55.
[122]周大镯.多变量时间序列的聚类、相似查询与异常检测[D].天津:天津大学,2008,12.
[123] Hawkins D..Identification of outliers[M]. London: Chapman&Hall,1980.
[124] Ye Jun. Cosine similarity measures for intuitionistic fuzzy sets and their applications[J].Mathematical&Computer Modeling,2011,53(1-2):91-97.
[125] Zhu Shiwei, Wu Junjie, Xiong Hui, Xia Guoping. Scaling up top-K cosine similaritysearch[J]. Data&Knowledge Engineering,2011,70:60-83.
[126] Liu Chengjun. Clarification of assumptions in the relationship between the Bayes decisionrule and the whitened cosine similarity measure[J]. IEEE Transactions on Pattern Analysis&Machine Intelligence,2008,30(6):1116-1117.
[127] Sergios T., Konstantinos K.著,李晶皎,王爱侠,张广渊等译.模式识别(第4版)[M].北京:电子工业出版社,2010,2.
[128]王钰,周志华,周傲英.机器学习及其应用[M].北京:清华大学出版社,2006,3.
[129] Cover T.M., Hart P.E.. Nearest neighbor pattern classification[J]. IEEE Transaction onInformation Theory,1967,6:113-121.
[130]陆婉珍.现代近红外光谱分析技术(第2版)[M].北京:中国石化出版社,2007,1.
[131]同济大学数学系.高等数学(第6版,上册)[M].北京:高等教育出版社,2007.
[132] Geladi P., Macdougall D., Martens H.. Linearization and scatter-correction for near-infraredreflectance spectra of meat[J]. Applied Spectroscopy,1985,39(3):491-500.
[133] Isaksson T., Kowalski B.R.. Piece-wise multiplicative scatter correction applied tonear-infrared transmittance data from meat products[J]. Applied Spectroscopy,1993,47(6):702-709.
[134] Windig W., Shaver J., Bro R.. Loopy MSC: a simple way to improve multiplicative scattercorrection[J]. Applied Spectroscopy,2008,62(10):1153-1159.
[135]芦永军,曲艳玲,宋敏.近红外相关光谱的多元散射校正处理研究[J].光谱学与光谱分析,2007,27(5):877-880.
[136] Maleki M.R., Mouazen A.M., Ramon H., Baerdemaeker J.D.. Multiplicative scattercorrection during on-line measurement with near infrared spectroscopy[J]. BiosystemsEngineering,2007,96(3):427-433.
[137] Thennadil S.N., Martens H., Kohler A.. Physics-based multiplicative scatter correctionapproaches for improving the performance of calibration models[J]. Applied Spectroscopy,2006,60(3):315-321.
[138] Chen J.Y., Iyo C.. Effect of multiplicative scatter correction on wavelength selection for nearinfrared calibration to determine fat content in raw milk[J]. Journal of Near InfraredSpectroscopy,2002,10(4):301-307.
[139]靳刘蕊.函数性数据分析方法及应用研究[D].厦门:厦门大学,2008,3.
[140]梁逸曾,俞汝勤.分析化学手册(10)——化学计量学[M].北京:化工出版社,2001.
[141] Tao Pan, Jiang Guoqiang, Chen Jiemei. Waveband Selection of NIR Spectroscopy Analysisfor Glucose Aqueous Solution Based on Savitzky-Golay Smoothing[J]. Advanced MaterialsResearch,2011,181-182:712-716.
[142] Delwiche S.R., Reeves J.B.. A graphical method to evaluate spectral preprocessing inmultivariate regression calibrations: Example with Savitzky-Golay filters and partial leastsquares regression[J]. Applied Spectroscopy,2010,64(1):73-82.
[143]谢军,潘涛,陈洁梅,陈华舟,任小焕.血糖近红外光谱分析的Savitzky-Golay平滑模式与偏最小二乘法因子数的联合优选[J].分析化学,2010,38(3):342-346.
[144]褚小立,袁洪福,陆婉珍.近红外分析中光谱预处理及波长选择方法进展与应用[J].化学进展,2004,16(4):528-541.
[145] Savitzky A., Golay M.J.E.. Smoothing and differentiation of data by simplified least squaresprocedures[J]. Analytical Chemometrics,1964,36(8):1627-1639.
[146] Gorry P.A.. General least-squares smoothing and differentiation by the convolution(Savitzky-Golay) method [J]. Analytical Chemometrics,1990,62:570-573.
[147] Huang L.H.. A roughness-penalty view of kernel smoothing[J]. Statistics&ProbabilityLetters,2001,52(1):85-89.
[148] Kennedy J., Eberhart R.C.. Particle swarm optimization[C]. Proceedings of The IEEEInternational Conference on Neural Networks,1995,4:1942-1948.
[149] Shi Y., Eberhart R.C.. A modified particle swarm optimizer[C]. Proceedings of The IEEEInternational Conference on Evolutionary Computation,1998:69-73.
[150] Eberhart R.C., Shi Y.H.. Particle swarm optimization: Developments, applications andresources[C]. Proceedings of The2001Congress on Evolutionary Computation,2001,1-2:81-86.
[151] Ethem A.著,范明译.机器学习导论[M].北京:机械工业出版社,2009,6.
[152] Blanchet F.G., Legendre P., Borcard D.. Forward selection of explanatory variables[J].Ecology,2008,89(9):2623-2632.
[153] Whitley D.C., Ford M.G., Livingstone D.J.. Unsupervised forward selection: A method foreliminating redundant variables[J]. Journal of Chemical Information&Computer Sciences,2000,40(5):1160-1168.
[154]徐云.农产品品质检测中的近红外光谱分析方法研究[D].北京:中国农业大学,2009.
[155] Holland J. H.. Adaptation in nature and artificial systems[M]. Michigan: The University ofMichigan Press,1975.
[156] Bagley J. D.. The behavior of adaptive system which employs genetic and correlationalgorithm[M]. Michigan: The University of Michigan Press,1967.
[157] Goldberg D.. Genetic algorithm in search, optimization and machine learning[M]. Boston:Addison Wesley Longman Publishing Co.,1989.
[158] Davis L.. Handbook of genetic algorithms[M]. New York: Van Nostrand Reinhold,1991.
[159]张顶学.遗传算法与粒子群算法的改进及应用[D].武汉:华中科技大学,2007.
[160] Fei Qiang, Li Ming, Wang Bin, Huan Yanfu, Feng Guodong, Ren Yulin. Analysis ofcefalexin with NIR spectrometry coupled to artificial neural networks with modified geneticalgorithm for wavelength selection[J]. Chemometrics&Intelligent Laboratory Systems,2009,97(2):127-131.
[161] Xu Zhibin, Yu Lei, Lei Meng, An Baoran. Application research on coal analysis of nearinfrared spectroscopy (NIRS) by intelligent algorithms[C]. Proceedings of The22nd ChineseControl&Decision Conference,2010,2416-2419.
[162] Shi Bolin, Ji Baoping, Zhu Dazhou, Tu Zhenhua, Qing Zhaoshen. Study on geneticalgorithms-based nir wavelength selection for determination of soluble solids content in fujiapples[J]. Journal of Food Quality,2008,31(2):232-249.
[163] Reynes Ch., de Souza S., Sabatier R., Figueres G., Vidal B.. Selection of discriminantwavelength intervals in NIR spectrometry with genetic algorithms[J]. Journal ofChemometrics,2006,20(3-4):136-145.
[164] Wu Jianning, Wang Jue, Liu Li. Feature extraction via KPCA for classification of gaitpatterns[J]. Human movement science,2007,26:393-411.
[165] Michael B.R., Indra A.. Classification and regionalization through kernel principalcomponent analysis[J]. Physics&Chemistry of The Earth,2010,35:316-328.
[166] Fu G., Frank Y. Shih, Wang H.M.. A kernel-based parametric method for conditional densityestimation[J]. Pattern Recognition,2011,44:284-294.
[167] Zhao Lihong, Zhang Xili, Xu Xinhe. Face recognition base on KPCA with polynomialkernels[C]. Proceedings of The5th International Conference on Wavelet Analysis&PatternRecognition,2007,1-4:1213-1216.
[168] Jorgensen K.W., Hansen L.K.. Model selection for gaussian kernel PCA denoising[J]. IEEETransactions on Neural Networks&Learning Systems,2012,23(1):163-168.
[169] Kohonen, T.. Essentials of the self-organizing map[J]. Neural Networks,2013,37:52-65.
[170] Vesanto J., Alhoniemi E.. Clustering of the self-organizing map[J]. IEEE Transactions onNeural Networks,2000,11(3):586-600.
[171] Creput J.C., Hajjam A., Koukam A., Kuhn O.. Self-organizing maps in population basedmetaheuristic to the dynamic vehicle routing problem[J]. Journal of CombinatorialOptimization,2012,24(4):437-458.
[172] Shieh S., Liao I.. A new approach for data clustering and visualization using self-organizingmaps[J]. Expert Systems with Applications,2012.39(15):11924-11933.
[173] Ballabio D., Vasighi M.. A MATLAB toolbox for self organizing maps and supervised neuralnetwork learning strategies[J]. Chemometrics&Intelligent Laboratory Systems,2012,118:24-32.
[174] Hong Yingyi, Chou Jinghan. Nonintrusive energy monitoring for microgrids using hybridself-organizing feature-mapping networks[J]. Energies,2012,5(7):2578-2593.
[175] Huang G.B., Saratchandran P., Sundararajan N.. A generalized growing and pruning RBF(GGAP-RBF) neural network for function approximation[J]. IEEE Transactions on NeuralNetworks,16(1):57-67.
[176] Telmoudi A.J., Tlijani H., Nabli L., Ali M., M'Hiri R.. A new RBF neural network forprediction in industrial control[J]. International Journal of Information Technology&Decision Making,2012,11(4):749-775.
[177] She Yuanguo. Hybrid intelligent algorithm of RBF neural network with variable lengthstructure[J]. Information-An International Interdisciplinary Journal,2012,15(6):2571-2577.
[178] Han Honggui, Qiao Junfei. Adaptive computation algorithm for RBF neural network[J].IEEE Transactions on Neural Networks&Learning Systems,2012,23(2):342-347.
[179] Du Dajun, Li Kang, Fei Minrui. A fast multi-output RBF neural network constructionmethod[J]. Neurocomputing,2010,73(10-12):2196-2202.
[180] Lin C.L., Wang J.F., Chen C.Y., Chen C.W., Yen C.W.. Improving the generalizationperformance of RBF neural networks using a linear regression technique[J]. Expert Systemswith Applications,2009,36(10):12049-12053.
[181] Georgia M., Antreas A., Kalliopi M., Haralambos S., Olga I.M.. Prediction of toxicity usinga novel RBF neural network training methodology[J]. Journal of Molecular Modeling,2006,12(3):297-305.
[182] Dietterich T.G.. Machine Leaning Research: Four Current Directions[J]. AI Magazine,1997,18(4):97-136.
[183] Hansen L.K., Salamon P.. Neural Network Ensembles[J]. IEEE Transactions on PatternAnalysis&Machine Intelligence,1990,12(10):993-1001.
[184] Ahn B.S., Cho S.S., Kim C.Y.. The integrated methodology of rough set theory and artificialneural network for business failure prediction[J]. Expert Systems with Applications,2000,18(2):65-74.
[185] Chen J.. A predictive system for blast furnaces by integrating a neural network withqualitative analysis[J]. Engineering Applications of Artificial Intelligence,2001,14(1):77-85.
[186]沈掌泉.神经网络集成技术及其在土壤学中应用的研究[D].杭州:浙江大学,2005,5.
[187] Dor O., Zhou Y.Q.. Real-SPINE: An integrated system of neural networks for real-valueprediction of protein structural properties[J]. Proteins-Structure Function&Bioinformatics,2007,68(1):76-81.
[188] Xue B., Faraggi E., Zhou Y.. Predicting residue-residue contact maps by a two-layer,integrated neural-network method[J]. Proteins-Structure Function&Bioinformatics,2009,76(1):176-183.
[189] Azadeh A., Maghsoudi A., Sohrabkhani S.. An integrated artificial neural networks approachfor predicting global radiation[J]. Energy Conversion&Management,2009,50(6):1497-1505.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |