View in EDS

Machine learning and design of experiment for enhanced biomass using N2 and CO2-enriched biochar at different pyrolysis temperatures.

Saved in:
Bibliographic Details
Title: Machine learning and design of experiment for enhanced biomass using N2 and CO2-enriched biochar at different pyrolysis temperatures.
Authors: Aasim, Muhammad1 (AUTHOR) mshazim@gmail.com, Aydin, Ebubekir Siddik2 (AUTHOR) esaydin@sivas.edu.tr, Korkut, Ibrahim2 (AUTHOR) ikorkut@sivas.edu.tr, Ergan, Başak Temur3 (AUTHOR) btemur@gtu.edu.tr, Say, Ahmet4 (AUTHOR) ahmetsay@erciyes.edu.tr, Soomro, Salma Naimatullah5 (AUTHOR) sn_soomro@hotmail.com, Soomro, Sabeen Rehman5 (AUTHOR) sabeen_rehman@yahoo.com, Ali, Seyid Amjad6 (AUTHOR) syedali@bilkent.edu.tr, Ercişli, Sezai7 (AUTHOR) sercisli@gmail.com, Katırcı, Ramazan8 (AUTHOR) ramazank2000@gmail.com
Source: Biomass & Bioenergy. May2026, Vol. 208, pN.PAG-N.PAG. 1p.
Subjects: Biochar, Machine learning, Aquatic plants, Response surfaces (Statistics), Prediction models, Experimental design, Plant biomass
Abstract: This study presents the impact of five different biochars prepared at different pyrolysis temperatures, enriched with nitrogen and carbon dioxide as a precursor. The study was developed by utilizing design of experiment (DOE) for response surface methodology (RSM). Different concentrations of biochar were added to the liquid culture medium for in vitro regeneration of Ceratophyllum demersum , which revealed the better performance of biochar prepared at low temperatures on plant biomass and pigmentation. Results analyzed by correlation analysis exhibited a significantly negative impact of biochar concentration and type on plant biomass and biochemical parameters, underscoring the intricate relationship within plant antioxidative mechanisms and growth response. Different machine learning (ML) models namely Random Forest (RF), Multi-Layer Perceptron (MLP), Gaussian Process Regression (GPR), Extreme Gradient Boosting (EXGB), CatBoost Regressor (CBR), and LightGBM Regressor (LGBMR) were employed for predictive accuracy and robustness using the coefficient of determination (R2) and Mean Squared Error (MSE). Results indicated that RF, XGB, and CBR consistently achieved superior predictive performance for chlorophyll-a (Chl-a), chlorophyll-b (Chl-b), and total chlorophyll (Chl-T), tannic acid (TA), with high R2 and low MSE values. The RF model used for the Y-Permutation test, feature importance test, and SHAP test confirmed the predictive performance. Feature importance test figured out concentration, while SHAP analysis illustrated biochar type as the most significant factor despite close scores. These insights may help to formulate next-generation biochar production for sustainable crop production and wastewater-based plant systems under climate stress conditions. [Display omitted] • Key Highlights. • Pyrolysis-derived biochars enriched with CO 2 or N 2 offer tunable properties for plant systems. • Response Surface Methodology (RSM) enables efficient optimization of biochar application. • Ensemble machine learning models enhance predictive accuracy for plant physiological traits. • First study on Integrated ML–RSM frameworks for in vitro biochar–plant interactions. [ABSTRACT FROM AUTHOR]
Copyright of Biomass & Bioenergy is the property of Pergamon Press - An Imprint of Elsevier Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:This study presents the impact of five different biochars prepared at different pyrolysis temperatures, enriched with nitrogen and carbon dioxide as a precursor. The study was developed by utilizing design of experiment (DOE) for response surface methodology (RSM). Different concentrations of biochar were added to the liquid culture medium for in vitro regeneration of Ceratophyllum demersum , which revealed the better performance of biochar prepared at low temperatures on plant biomass and pigmentation. Results analyzed by correlation analysis exhibited a significantly negative impact of biochar concentration and type on plant biomass and biochemical parameters, underscoring the intricate relationship within plant antioxidative mechanisms and growth response. Different machine learning (ML) models namely Random Forest (RF), Multi-Layer Perceptron (MLP), Gaussian Process Regression (GPR), Extreme Gradient Boosting (EXGB), CatBoost Regressor (CBR), and LightGBM Regressor (LGBMR) were employed for predictive accuracy and robustness using the coefficient of determination (R2) and Mean Squared Error (MSE). Results indicated that RF, XGB, and CBR consistently achieved superior predictive performance for chlorophyll-a (Chl-a), chlorophyll-b (Chl-b), and total chlorophyll (Chl-T), tannic acid (TA), with high R2 and low MSE values. The RF model used for the Y-Permutation test, feature importance test, and SHAP test confirmed the predictive performance. Feature importance test figured out concentration, while SHAP analysis illustrated biochar type as the most significant factor despite close scores. These insights may help to formulate next-generation biochar production for sustainable crop production and wastewater-based plant systems under climate stress conditions. [Display omitted] • Key Highlights. • Pyrolysis-derived biochars enriched with CO 2 or N 2 offer tunable properties for plant systems. • Response Surface Methodology (RSM) enables efficient optimization of biochar application. • Ensemble machine learning models enhance predictive accuracy for plant physiological traits. • First study on Integrated ML–RSM frameworks for in vitro biochar–plant interactions. [ABSTRACT FROM AUTHOR]
ISSN:09619534
DOI:10.1016/j.biombioe.2025.108803