Recently, a student from ICARE has published a paper entitled Machine learning prediction of pyrolytic gas yield and compositions with feature reduction methods: Effects of pyrolysis conditions and biomass characteristics on Bioresource Technology (link: https://doi.org/10.1016/j.biortech.2021.125581). The host institution of the first author is Huazhong University of Science and Technology. The first author is Mr. Tang Qinghui, a graduate student of ICARE. The corresponding authors are Professor Yang Haiping from the State Key Laboratory of Coal Combustion of Huazhong University of Science and Technology, and Professor Wang Shurong from the State Key Laboratory of Clean Energy Utilization of Zhejiang University.
The study of this paper aimed to utilize machine learning algorithms combined with feature reduction for predicting pyrolytic gas yield and compositions based on pyrolysis conditions and biomass characteristics. To this end, random forest (RF) and support vector machine (SVM) were introduced and compared. The results suggested that six features were adequate to accurately forecast (R2 > 0.85, RMSE < 5.7%) the yield while the compositions only required three. Moreover, the profound information behind the models was extracted. The relative contribution of pyrolysis conditions was higher than that of biomass characteristics for yield (55%), CO2 (73%), and H2 (81%), which was inverse for CO (12%) and CH4 (38%). Furthermore, partial dependence analysis quantified the effects of both reduced features and their interactions exerted on the pyrolysis process. This study provided references for pyrolytic gas production and upgrading in a more convenient manner with fewer features, and extended the knowledge into the biomass pyrolysis process.
Professor Yang Haiping, a Chinese superior of ICARE is the co-corresponding author of the paper. She is also a professor and Ph.D supervisor of Huazhong University of Science and Technology, winner of the National Science Fund for Distinguished Young Scholars. She was selected as Elsevier China's highly cited scholar for seven consecutive years from 2014 to 2020, and has won the Young Scientist Fund of the National Natural Science Fund (China) in 2016 and the Newton Advanced Fellowship of the Royal Society(Great Britain) in 2018. She has publications in numerous academic journals such as Combustion and Flame, Applied Catalysis B: Environmental, Fuel, Energy and Fuels, Fuel Processing Technology, and Bioresources Technology. She is currently the Deputy Editor of Fuel Processing Technology, and the editorial board member of Fuel, Journal of the Energy Institute and Journal of analytical and applied pyrolysis.
Mr. Tang Qinghui, the first author of the paper, is a master student of ICARE. He won the Excellent Graduate Student award in 2020, went to Paris University of Letters and Sciences, France for one year’s joint training, and won the National Scholarship in 2021. At present, he focuses his research on the combination of biomass pyrolysis and machine learning. As the first author, he has already published two academic papers in Energy & Fuels and Bioresource Technology.
