Abstract
Prediction of pulp yield of Eucalyptus globulus wood samples based on partial least squares (PLS) regression can be optimized by utilizing specific near infrared (NIR) wavelengths. A critical feature of this approach is the weighting of constraint conditions. Equal weighting balances optimization in terms of calibration and prediction; however, there is a lack of knowledge regarding prediction performance of wood property models when different weight factors are used. In this study, pulp yield models were developed using two E. globulus data sets characterized by narrow (5%) and extreme (22.6%) yield ranges and represented by untreated and second derivative NIR spectra. The global optimization solver pySOT was used to optimize the performance of a PLS regression model in terms of wavelengths selected and number of latent variables. A linear function of R-squares for calibration (\({R}_{c}^{2}\)) and prediction (\({R}_{p}^{2}\)) sets was utilized as the objective function with the aim of maximizing \(\alpha {R}_{c}^{2}+\left(1-\alpha \right){R}_{p}^{2}\) for all values of \(\alpha\) between 0 (maximizing \({R}_{p}^{2}\) without concern for \({R}_{c}^{2}\)) and 1 (only maximizing \({R}_{c}^{2}\)). Values of \(\alpha \le 0.8\) provided good predictive performance, whereas \(\alpha \ge 0.9\) tended to overfit the calibration data indicating that models are robust for values of \(\alpha\) from 0 to 0.8. Representative wavelengths for each data set were identified and assigned to corresponding wood components through a band assignment process. Strong agreement was observed for \(\alpha \le 0.8\); however, for \(\alpha \ge 0.9,\) identified wavelengths generally occurred in regions unrelated to vibrations arising from specific wood components.
Similar content being viewed by others
References
Ayanleye S, Nasir V, Avramidis S, Cool J (2021) Effect of wood surface roughness on prediction of structural timber properties by infrared spectroscopy using ANFIS, ANN and PLS regression. Eur J Wood Prod 79(1):101–115
Bangalore AS, Shaffer RE, Small GW, Arnold MA (1996) Genetic algorithm -based method for selecting wavelengths and model size for use with partial least-squares regression: application to near-infrared spectroscopy. Anal Chem 68(23):4200–4212
Cogdill RP, Schimleck LR, Jones PD, Peter GF, Daniels RF, Clark A (2004) Estimation of the physical wood properties of Pinus taeda L. radial strips using least squares support vector machines. J Near Infrared Spectrosc 12(4):263–270
De A, Chanda S, Tudu B, Bandyopadhyay RB, Hazarika AK, Sabhapondit S, Baruah BD, Tamuly P, Bhattachryya N (2017) Wavelength Selection for Prediction of Polyphenol Content in Inward Tea Leaves Using NIR. In: IEEE 7th international advance computing conference (IACC), Hyderabad, 2017 pp 184–187
Downes GM, Meder R, Bond H, Ebdon N, Hicks C, Harwood C (2011) Measurement of cellulose content, Kraft pulp yield and basic density in eucalypt woodmeal using multisite and multispecies near infra-red spectroscopic calibrations. South for 73(3–4):181–186
Downes GM, Worledge D, Schimleck LR, Harwood C, French J, Beadle CL (2006) The effect of growth rate and irrigation on the basic density and kraft pulp yield of Eucalyptus globulus and E. nitens. N Z J For 51(3):13–22
Ehrgott M (2005) Multicriteria Optimization. Springer, Berlin Heidelberg, Germany
Eldridge KG, Davidson J, Harwood CE, vanWyk G (1993) Eucalypt domestication and breeding. Oxford University Press, Oxford
Eriksson D, Bindel D, Shoemaker CA (2015) Surrogate optimization toolbox (pySOT) (2015) Available from https://github.com/dme65/pySOT
Eriksson D, Bindel D, Shoemaker CA (2019) pySOT and POAP: An event-driven asynchronous framework for surrogate optimization. ArXiv, abs/1908.00420
Evans R (1994) Rapid measurement of the transverse dimensions of tracheids in radial wood sections from Pinus radiata. Holzforschung 48:168–172
Evans R (1999) A variance approach to the X-ray diffractometric estimation of microfibril angle in wood. Appita J 52(283–289):294
Evans R (2006) Characterization of the cellulosic cell wall. Stokke DG, Groom L (ed) pp 138–146. Blackwell Publishing, Ames, IA, USA
Fernandes A, Lousada J, Morais J, Xavier J, Pereira J, Melo-Pinto P (2013) Measurement of intra-ring wood density by means of imaging VIS/NIR spectroscopy (hyperspectral imaging). Holzforschung 67(1):59–65
Greaves BL, Borralho NMG (1996) The influence of basic density and pulp yield on the cost of eucalypt kraft pulping: A theoretical model for tree breeding. Appita J 49(2):90–95
Ho TX, Schimleck LR, Sinha A (2021) Utilization of genetic algorithms to optimize Eucalyptus globulus pulp yield models based on NIR spectra. Wood Sci Technol 55(3):757–776
Ho TX, Schimleck LR, Sinha A, Dahlen J (2022) Utilization of genetic algorithms to optimize loblolly pine wood property models based on NIR spectra and SilviScan data. Wood Sci Technol 56:1419–1437. https://doi.org/10.1007/s00226-022-01403-z
Li Y, Via BK, Cheng Q, Zhao J, Li Y (2019) New pretreatment methods for visible–near-infrared calibration modeling of air-dry density of Ulmus pumila wood. For Prod J 69(3):188–194
Michell AJ, Schimleck LR (1996) NIR spectroscopy of woods from Eucalyptus globulus. Appita J 49(1):23–26
Mora C, Schimleck LR (2010) Kernel regression methods for the prediction of wood properties of Pinus taeda using near infrared (NIR) spectroscopy. Wood Sci Technol 44(4):561–578
Nasir V, Nourian S, Zhou Z, Rahimi S, Avramidis S, Cool J (2019) Classification and characterization of thermally modified timber using visible and near-infrared spectroscopy and artificial neural networks: a comparative study on the performance of different NDE methods and ANNs. Wood Sci Technol 53(5):1093–1109
Raymond CA, Schimleck LR, Muneri A, Michell AJ (2001) Genetic parameters and genotype-by-environment interactions for pulp yield and pulp productivity in Eucalyptus globulus predicted using near infrared reflectance analysis. For Genet 8(3):213–224
Regis RG, Shoemaker CA (2007) A stochastic radial basis function method for the global optimization of expensive functions. INFORMS J Comput 19(4):497–509
Regis RG, Shoemaker CA (2013) Combining radial basis function surrogates and dynamic coordinate search in high-dimensional expensive black-box optimization. Eng Optim 45(5):529–555
Schimleck L, Apiolaza L, Dahlen J, Downes G, Emms G, Evans R, Moore J, Pâques L, Van den Bulcke J, Wang X (2019) Non-destructive evaluation techniques and what they tell us about wood property variation. Forests 10:728
Schimleck LR (2008) Near infrared spectroscopy: A rapid, non-destructive method for measuring wood properties and its application to tree breeding. N Z J for Sci 38(1):14–35
Schimleck LR, Tsuchikawa S (2021) Application of NIR spectroscopy to wood and wood derived products (Chapter 37). In: Ciurczak E, Igne B, Workman J, Burns D (eds) The handbook of near-infrared analysis, fourth edition, newly revised and expanded. CRC Press, Boca Raton, FL, pp 759–780
Schwanninger M, Rodrigues JC, Fackler K (2011) A review of band assignments in near infrared spectra of wood and wood components. J near Infrared Spectrosc 19:287–308
Snee R (1977) Validation of regression models: methods and examples. Technometrics 19:415–428
Trung T, Downes G, Meder R, Allison B (2015) Pulp mill and chemical recovery control with advanced analysers - from trees to final product. Appita J 68(1):39–46
Tsuchikawa S, Kobori H (2015) A review of recent application of near infrared spectroscopy to wood science and technology. J Wood Sci 61(3):213–220
Villar A, Fernandez S, Gorritxategi E, Ciria JI, Fernandez LA (2014) Optimization of the multivariate calibration of a Vis-NIR sensor for the on-line monitoring of marine diesel engine lubricating oil by variable selection methods. Chemometr Intell Lab Syst 130:68–75
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhen, Y., Ho, T.X., Roberts, L. et al. On the selection of the weighting parameter value in optimizing Eucalyptus globulus pulp yield models based on NIR spectra. Wood Sci Technol 56, 1835–1850 (2022). https://doi.org/10.1007/s00226-022-01431-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00226-022-01431-9