期刊文献+

木材抗拉强度的近红外光谱MC-UVE-IVSO建模方法 被引量:1

Near Infrared Spectroscopy Modeling Method of Wood Tensile Strength Based on MC-UVE-IVSO
下载PDF
导出
摘要 木材抗拉强度是评价木材力学性质的重要指标。针对近红外光谱建模中样本数据量小、波长信息冗余所导致预测模型精度低的问题,提出一种基于模型集群分析MC-UVE-IVSO波长优选的木材抗拉强度建模方法。以桦木为例,选取150个桦木样本作为实验对象,首先使用900~1700 nm波段的近红外光谱仪采集试件光谱数据,并采用力学试验机获得相应的抗拉强度真值;然后对采集的光谱数据运用多元散射校正(MSC)、一阶求导和卷积平滑(SG)相结合的方法进行预处理,完成光谱平滑滤波;分别采用变量组合集群分析算法(VCPA)、蒙特卡罗无信息变量消除法(MC-UVE)、迭代变量子集优化算法(IVSO)及MC-UVE-IVSO组合优化算法进行波长筛选,并对比优选波长结果;最后在优选近红外波长基础上,建立桦木抗拉强度的偏最小二乘预测模型(PLS)。实验结果表明:基于MC-UVE-IVSO算法优选波长的PLS模型,光谱变量数由512减小到98,优选波长占总波长的19%,其预测决定系数R^(2)为0.94,预测均方根误差RMSEP为7.50,性能偏差比RPD为3.16,相比于全波段、MC-UVE、VCPA、MC-UVE-VCPA与IVSO相应的R^(2)(0.92、0.93、0.82、0.87、0.93)、RMSEP(17.91、11.7、14.91、12.12、8.47)和RPD(2.81、2.91、2.25、2.28、2.78)均有不同程度提升;通过统计特征波长所建立的预测模型箱形图,进一步证明了MC-UVE-IVSO算法在处理多变量波长的稳定性。实验结果表明,MC-UVE方法可以消除与建模不相关的多数变量,而IVSO算法能有效搜索出最优变量子集,基于MC-UVE-IVSO的光谱优选算法提升了木材抗拉强度预测模型的准确性和稳定性,为木材近红外光谱的无损、快速与精准检测提供了一定的理论基础。 The tensile strength is an important index to assess the mechanical properties of the wood.In order to solve the problems of low model accuracy caused by the small samples and redundant wavelength information in near-infrared spectroscopy modeling,a novel method combining wavelength optimization of MC-UVE-IVSO and PLS is proposed to predict the wood tensile strength.Firstly,150 birch samples were selected as experimental objects,and the near-infrared spectrometer in the band of 900~1700 nm was used to collect the spectral data of the test specimens,and the true tensile strength values were obtained by the mechanical testing machine.Secondly,the collected spectral data were preprocessed to complete smoothing filtering by combining multivariate scattering correction(MSC),first-order derivation and convolution smoothing(SG).Thirdly,the optimization methods,which include the variable combination cluster analysis algorithm(VCPA),the Monte Carlo uninformative variable elimination method(MC-UVE),the iterative variable subset optimization algorithm(IVSO)and the MC-UVE-IVSO combined optimization algorithm,were applied to select spectral wavelength features,and the optimal wavelength results based on different method were compared.Finally,the partial least squares birch tensile strength prediction model was established based on the selected wavelength of MC-UVE-IVSO.The experimental results show that the number of spectral variables is reduced from 512 to 98 based on the MC-UVE-IVSO and PLS,and the selected wavelength features account for 19%of the total wavelength.The predicted coefficient of determination(R^(2))is 0.9404.The root mean square error of prediction(RMSEP)is 12.3707.The ratio of performance to deviation(RPD)is 3.1624,compared with full band,MC-UVE,VCPA,MC-UVE-VCPA and IVSO,R^(2)indicators(0.9265,0.8282,0.9317,0.9343),RMSEP indicators(13.9105,17.3552,13.4028,14.0705)and RPD indicators(2.8123,2.2541,2.9188,2.7803)have been improved to varying degrees;In addition,the box plot of the prediction model established by statistical characteristic wavelengths further proves the stability of the MC-UVE-IVSO algorithm in dealing with multivariate wavelengths.The experimental results proved that the MC-UVE method could eliminate most of the variables,which are not related to the model,and the IVSO algorithm can effectively search for the optimal subset of variables.Based on the MC-UVE-IVSO optimization algorithm,the combination method has complementary advantages,and the optimized features can improve the accuracy and stability of the birch tensile strength prediction model.The method provides a theoretical basis for Non-destructive testing of wood samples based on near-infrared spectroscopy.
作者 蒋大鹏 高礼彬 陈金浩 张怡卓 JIANG Da-peng;GAO Li-bin;CHEN Jin-hao;ZHANG Yi-zhuo(College of Computer Science and Artificial Intelligence,Changzhou University,Changzhou 213164,China;College of Mechanical and Electrical Engineering,Northeast Forestry University,Harbin 150040,China)
出处 《光谱学与光谱分析》 SCIE EI CAS CSCD 北大核心 2023年第8期2488-2493,共6页 Spectroscopy and Spectral Analysis
基金 国家自然科学基金项目(31700643) 林业公益性行业科研专项(201304510)资助。
关键词 木材抗拉强度 近红外光谱 集群分析 蒙特卡罗无信息变量消除 迭代变量子集优化 Wood tensile strength Near-infrared spectroscopy Model population analysis Monte Carlo uninformative variable elimination Iterative variable subset optimization
  • 相关文献

参考文献4

二级参考文献69

  • 1李庆波,张广军,徐可欣,汪曣.DS算法在近红外光谱多元校正模型传递中的应用[J].光谱学与光谱分析,2007,27(5):873-876. 被引量:24
  • 2梁逸曾, 许青松. 复杂体系仪器分析─白、灰、黑分析体系及其化学计量学算法. 北京: 化学工业出版社, 2013.
  • 3Cawley G C, Talbot N L C. Bioinformatics, 2006, 22(19): 2348-2355.
  • 4Chen T, Martin E. Anal. Chim. Acta., 2009, 631(1): 13-21.
  • 5Wang Q, Li H D, Xu Q S, Liang Y Z. Analyst, 2011, 136(7): 1456-1463.
  • 6Yeung K Y, Bumgarner R E, Raftery A E. Bioinformatics, 2005, 21(10): 2394-2402.
  • 7Candes E, Tao T. Ann. Stat., 2007: 2313-2351.
  • 8Johnstone I M, Titterington D M. Philos. Trans. A. Math. Phys. Eng. Sci., 2009, 367(1906): 4237-4253.
  • 9Zou H, Hastie T. J. Roy. Stat. Soc. B., 2005, 67(2): 301-320.
  • 10Li H D, Liang Y Z, Xu Q S, Cao D S. J. Chemometr, 2010, 24: 418-423.

共引文献40

同被引文献15

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部