Auto-classification of biomass through characterization of their pyrolysis behaviors using thermogravimetric analysis with support vector machine algorithm: case study for tobacco

Background During the biomass-to-bio-oil conversion process, many studies focus on studying the association between biomass and bio-products using near-infrared spectra (NIR) and chemical analysis methods. However, the characterization of biomass pyrolysis behaviors using thermogravimetric analysis (TGA) with support vector machine (SVM) algorithm has not been reported. In this study, tobacco was chosen as the object for biomass, because the cigarette smoke (including water, tar, and gases) released by tobacco pyrolysis reactions decides the sensory quality, which is similar to biomass as a renewable resource through the pyrolysis process. Results SVM algorithm has been employed to automatically classify the planting area and growing position of tobacco leaves using thermogravimetric analysis data as the information source for the first time. Eighty-eight single-grade tobacco samples belonging to four grades and eight categories were split into the training, validation, and blind testing sets. Our model showed excellent performances in both the training and validation set as well as in the blind test, with accuracy over 91.67%. Throughout the whole dataset of 88 samples, our model not only provides precise results on the planting area of tobacco leave, but also accurately distinguishes the major grades among the upper, lower, and middle positions. The error only occurs in the classification of subgrades of the middle position. Conclusions From the case study of tobacco, our results validated the feasibility of using TGA with SVM algorithm as an objective and fast method for auto-classification of tobacco planting area and growing position. In view of the high similarity between tobacco and other biomasses in the compositions and pyrolysis behaviors, this new protocol, which couples the TGA data with SVM algorithm, can potentially be extrapolated to the auto-classification of other biomass types.

[1]  J. Bilbao,et al.  Fast characterization of biomass fuels by thermogravimetric analysis (TGA) , 2015 .

[2]  Publisher's Note , 2018, Anaesthesia.

[3]  L. Tognotti,et al.  A Generalized Procedure for the Devolatilization of Biomass Fuels Based on the Chemical Components , 2014 .

[4]  Mohammad. Rasul,et al.  Biofuels Production through Biomass Pyrolysis —A Technological Review , 2012 .

[5]  Yun Xu,et al.  On Splitting Training and Validation Set: A Comparative Study of Cross-Validation, Bootstrap and Systematic Sampling for Estimating the Generalization Performance of Supervised Learning , 2018, Journal of Analysis and Testing.

[6]  E. Jakab,et al.  Tobacco pyrolysis. Kinetic evaluation of thermogravimetric–mass spectrometric experiments , 2009 .

[7]  Ran Li,et al.  Characterization of tobacco stalk lignin using nuclear magnetic resonance spectrometry and its pyrolysis behavior at different temperatures , 2019, Journal of Analytical and Applied Pyrolysis.

[8]  Taian Luo,et al.  Co-pyrolysis characteristics and kinetics of coal and plastic blends , 2009 .

[9]  M. Barker,et al.  Partial least squares for discrimination , 2003 .

[10]  Panagiotis Grammelis,et al.  Pyrolysis characteristics and kinetics of biomass residuals mixtures with lignite , 2003 .

[11]  D. T. Liang,et al.  In-Depth Investigation of Biomass Pyrolysis Based on Three Major Components: Hemicellulose, Cellulose and Lignin , 2006 .

[12]  M. Hajaligol,et al.  The vaporization of semi-volatile compounds during tobacco pyrolysis , 2006 .

[13]  R. Baker,et al.  The pyrolysis of tobacco ingredients , 2004 .

[14]  Fan Zhang,et al.  Classification and Quality Evaluation of Tobacco Leaves Based on Image Processing and Fuzzy Comprehensive Evaluation , 2011, Sensors.

[15]  Qi Song,et al.  Classification of the fragrant styles and evaluation of the aromatic quality of flue-cured tobacco leaves by machine-learning methods , 2016, J. Bioinform. Comput. Biol..

[16]  Weiming Yi,et al.  Comparative study on the pyrolysis of cellulose and its model compounds , 2019, Fuel Processing Technology.

[17]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[18]  Tamer Basar,et al.  The Structure of Dynamic Programming Processes , 2021, Dynamic Programming.

[19]  A. Riga,et al.  Thermogravimetric Analysis (TGA) , 2008 .

[20]  O. Olatunji,et al.  Property-based biomass feedstock grading using k-Nearest Neighbour technique , 2020 .

[21]  Pengxiang Zhao,et al.  Improving the accuracy of tree-level aboveground biomass equations with height classification at a large regional scale , 2013 .

[22]  P. Salatino,et al.  Patterns and kinetics of pyrolysis of tobacco under inert and oxidative conditions , 2007 .

[23]  Y. Sung,et al.  Thermogravimetric study on stem biomass of Nicotiana tabacum , 2009 .

[24]  Alberto Prieto,et al.  Computational intelligence and bioinspired systems , 2007, Neurocomputing.

[25]  P. A. Pilavachi,et al.  A study of lignocellulosic biomass pyrolysis via the pyrolysis of cellulose, hemicellulose and lignin , 2014 .

[26]  Nabil H. Farhat.  Photonic neural networks and learning machines , 1992, IEEE Expert.

[27]  L. A. Stone,et al.  Computer Aided Design of Experiments , 1969 .

[28]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[29]  Jie Zhang,et al.  Investigation of biomass torrefaction based on three major components: Hemicellulose, cellulose, and lignin , 2018, Energy Conversion and Management.

[30]  Haiping Yang,et al.  Characteristics of hemicellulose, cellulose and lignin pyrolysis , 2007 .

[31]  Xin Lu,et al.  A random forest of combined features in the classification of cut tobacco based on gas chromatography fingerprinting. , 2010, Talanta.

[32]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[33]  Jungho Im,et al.  Support vector machines in remote sensing: A review , 2011 .

[34]  Jun Xiao,et al.  Experimental Study of Biomass Pyrolysis Based on Three Major Components: Hemicellulose, Cellulose, and Lignin , 2011 .

[35]  Xiao Dong Chen,et al.  Study on the Effect of Cut Tobacco Drying on the Pyrolysis and Combustion Properties , 2014 .

[36]  Yi-xi Cai,et al.  Effects of the cellulose, xylan and lignin constituents on biomass pyrolysis characteristics and bio-oil composition using the Simplex Lattice Mixture Design method , 2017 .

[37]  Li-Jun Ni,et al.  Pattern recognition of Chinese flue-cured tobaccos by an improved and simplified K-nearest neighbors classification algorithm on near infrared spectra. , 2009, Analytica chimica acta.

[38]  Age K. Smilde,et al.  UvA-DARE ( Digital Academic Repository ) Assessment of PLSDA cross validation , 2008 .

[39]  Ronghou Liu,et al.  Kinetics and reaction chemistry of pyrolysis and combustion of tobacco waste , 2015 .

[40]  José L. Figueiredo,et al.  Pyrolysis kinetics of lignocellulosic materials—three independent reactions model , 1999 .

[41]  Malcolm Davidson,et al.  Forest classification and impact of BIOMASS resolution on forest area and aboveground biomass estimation , 2017, Int. J. Appl. Earth Obs. Geoinformation.

[42]  D. Bahler,et al.  Applying Artificial Neural Networks: Part II. Using near Infrared Data to Classify Tobacco Types and Identify Native Grown Tobacco , 1997 .

[43]  Oskar Faix,et al.  Thermogravimetry/mass spectrometry study of six lignins within the scope of an international round robin test , 1995 .

[44]  Michael Jerry Antal,et al.  Kinetic modeling of biomass pyrolysis , 1997 .

[45]  Danchen Zhu,et al.  Correlation of Feedstock and Bio-oil Compound Distribution , 2017 .

[46]  Lin Xie,et al.  Support Vector Machine Optimized by Genetic Algorithm for Data Analysis of Near-Infrared Spectroscopy Sensors , 2018, Italian National Conference on Sensors.

[47]  V. Pareek,et al.  Biomass pyrolysis—A review of modelling, process parameters and catalytic studies , 2015 .

[48]  Chia-Hua Ho,et al.  Recent Advances of Large-Scale Linear Classification , 2012, Proceedings of the IEEE.

[49]  H. Klus,et al.  Tobacco smoke: unraveling a controversial subject. , 2008, Experimental and toxicologic pathology : official journal of the Gesellschaft fur Toxikologische Pathologie.

[50]  M. Ferrão,et al.  Classification of biomass through their pyrolytic bio-oil composition using FTIR and PCA analysis , 2018 .