Prediction of Higher Heating Values in Bio-Oil from Solvothermal Biomass Conversion and Bio-Oil Upgrading Given Discontinuous Experimental Conditions

Abstract

Both the conversion of lignocellulosic biomass to bio-oil (BO) and the upgrading of BO have been the targets of many studies. Due to the large diversity and discontinuity seen in terms of reaction conditions, catalysts, solvents, and feedstock properties that have been used, a comparison across different publications is difficult. In this study, machine learning modeling is used for the prediction of final higher heating value (HHV) and ΔHHV for the conversion of lignocellulosic feedstocks to BO, and BO upgrading. The models achieved coefficient of determination (R²) scores ranging from 0.77 to 0.86, and the SHapley Additive exPlanations (SHAP) values were used to obtain model explainability, revealing that only a few experimental parameters are largely responsible for the outcome of the experiments. In particular, process temperature and reaction time were overwhelmingly responsible for the majority of the predictions, for both final HHV and ΔHHV. Elemental composition of the starting feedstock or BO dictated the upper possible HHV value obtained after the experiment, which is in line with what is known from previous methodologies for calculating HHV for fuels. Solvent used, initial moisture concentration in BO, and catalyst active phase showed low predicting power, within the context of the data set used. The results of this study highlight experimental conditions and variables that could be candidates for the creation of minimum reporting guidelines for future studies in such a way that machine learning can be fully harnessed.

Figure 1

Publication trend of studies related to bio-oil upgrading found in Web of Science using bio-oil upgrading as the search string.

Figure 2

Violin distribution and box-plots for elemental composition, reaction time, reaction temperature, original HHV, and change in HHV after processing for (a) BO and (b) SF.

Figure 3

Visual representation of the learning process of XGBoost.

Figure 4

Model performance for the final HHV and ΔHHV for lignocellulosic SF conversion to BO through solvolysis and BO upgrading. (a) Final HHV for solvolysis BO. (b) ΔHHV for solvolysis BO. (c) Final HHV for BO upgrading. (d) ΔHHV for BO upgrading.

Figure 5

SHAP values for (a) final HHV and (b) ΔHHV from solvolysis of lignocellulosic SF. SHAP values for (c) final HHV and (d) ΔHHV from BO upgrading.

Figure 6

Partial dependency plot for ΔHHV changes at different temperatures and reaction time values for solvolysis of biomass to BO in (a) and (b), with solvents ethanol, methanol, polyethylene (PEG)-glycerol, propanol, water, and water–ethanol mixture, and feedstocks divided into either lignin or lignocellulose. BO upgrading in (c) and (d), with solvents butanol, ethanol, methanol, propanol, water or no solvent, and the following feedstocks: cornstalk, duckweed, gumweed, juniper, oil palm empty fruit bunch, pubescens, rice husk, and wood.