Production and Prediction of The Higher Heating Value of Corncob Biochar Using Machine Learning Techniques
Article Sidebar
Main Article Content
Abstract
This study aims to enhance the quality of agricultural corncob biomass through the torrefaction process to improve its potential as a solid biofuel. The process was designed and simulated using Aspen Plus V12.1 and Design Expert V13.0, considering four main factors: biomass particle size, nitrogen gas flow rate, temperature, and reaction time. The response variable was the higher heating value (HHV). Experimental results revealed that the highest HHV of 19.3945 MJ/kg was obtained under the conditions of 20 mesh particle size, 30 mL/min nitrogen flow rate, 280°C, and 60 minutes, with an average deviation of 2.37%. Optimization using Design Expert indicated the optimal conditions at 57.60 mesh particle size, 27.40 mL/min nitrogen flow rate, 318.40 °C, and 66 minutes, yielding an HHV of 18.4853 MJ/kg. Machine learning techniques were employed to predict HHV, and the XGBoost (XGB) model demonstrated the highest prediction accuracy of 97.53%. Carbon content exhibited the strongest positive correlation with HHV, whereas oxygen and ash contents showed negative correlations. The results indicate that integrating process simulation with machine learning provides an effective approach to predict and optimize the torrefaction performance of corncob biomass, supporting the development of sustainable high-quality solid biofuel.
Article Details
References
ปนัดดา อินทร์ดำ, โชคชัย เหมือนมาศ, และสุกฤทธิรา รัตนวิไล. (2562).การคัดเลือกปัจจัยที่ส่งผลต่อกระบวนการทอร์ริแฟคชันของทะลายปาล์มโดยใช้การออกแบบการทดลองด้วย. วารสารวิจัยมหาวิทยาลัยขอนแก่น (ฉบับบัณฑิตศึกษา), 19(4), 86–99.
ธนศิษฏ์ วงศ์อำนวย. (2562, 26 ตุลาคม). การผลิตถ่านชีวภาพหรือไบโอชาร์ และการใช้ไบโอชาร์. มหาวิทยาลัยแม่โจ้. https://erp.mju.ac.th/acticleDetail.aspx?qid=1072.
Box, G. E. P., & Behnken, D. W. (1960). Some new three level designs for the study of quantitative variables. Technometrics, 2(4), 455–475. https://doi.org/10.1080/00401706.1960.10489912
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
Chen, Z., Pears, N., Freeman, M., & Austin, J. (2009). Road vehicle classification using support vector machines. IEEE International Conference on Intelligent Computing and Intelligent Systems, 2009,214–218. https://doi.org/10.1109/ICICISYS.2009.5357707
Kartal, F., & Ozveren, U. (2020). A deep learning approach for prediction of syngas lower heating valuefrom CFB gasifier in Aspen Plus. Energy, 209, 118457. https://doi.org/10.1016/j.energy.2020.118457
Katongtung, T., Onsree, T., & Tippayawong, N. (2022). Machine learning prediction of biocrude yields and higher heating values from hydrothermal liquefaction of wet biomass and wastes. Bioresource Technology, 344, 126278. https://doi.org/10.1016/j.biortech.2021.126278
Tuychiev, B. (2023, 23 February). Using XGBoost in Python Tutorial. https://www.datacamp.com/community/tutorials/xgboost-in-python
Nhuchhen, D. R., Basu, P., & Acharya, B. (2014). A comprehensive review on biomass torrefaction. International Journal of Renewable Energy & Biofuels, 2014(2014), 1–56. https://doi.org/10.5171/2014.506376
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, É. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.
Zaman, S. A., & Ghosh, S. (2021). A generic input-output approach in developing and optimizing an Aspen Plus steam-gasification model for biomass. Bioresource Technology, 337, 125412. https://doi.org/10.1016/j.biortech.2021.125412
