A novel waste-to-energy conversion plant based on catalytic pyrolytic conditions: Modeling and optimization using supervised machine learning and desirability-driven methodologies

Aqueel Ahmad, Ashok Kumar Yadav and Shifa Hasan

Energy, 2025, vol. 317, issue C

Abstract: The persistent nature of plastics significantly contributes to waste generation, making their disposal a critical issue. Pyrolysis, which converts plastic waste into fuel, is a promising solution to the energy crisis. This study evaluates the pyrolysis potential of low-density polyethylene (LDPE) plastic waste, focusing on the influence of process parameters (reactor temperature, residence time, N2 flow rate, and catalyst concentration) and their optimization using seven supervised machine learning models (RF, GB, ET, DT, KNN, AB, and XGB) and a desirability approach. Feature selection techniques (PCC, F-score, and MI) were employed to identify pertinent input features, with MI exhibiting superior performance in accurately identifying and combining the most relevant features for the model's predictive accuracy. The gradient boosting (GB) model demonstrated the highest predictive accuracy with R2 scores of 0.948, 0.9799, and 0.9802 for bio-oil, biochar, and syngas yields, respectively, and the lowest errors (RMSE: 0.6859, 0.1349, and 0.4194; MAE: 0.4505, 0.078, and 0.2673; MAPE: 0.681, 1.1379, and 0.9629). The tree-based pipeline optimization approach shows the optimal conditions of a reactor temperature of 450 °C, a residence time of 2 h, an N2 flow rate of 150 mL/min, and a catalyst concentration of 6 wt %, resulting in a maximum bio-oil yield of 72.20 % and biochar and syngas yields of 6.35 % and 23.48 %, respectively. Detailed characterization via GC-MS analysis indicated that bio-oil, which is composed of various organic compounds, such as phenols, ketones, alcohols, esters, and aromatic hydrocarbons, can be upgraded for use as a hydrocarbon fuel. Biochar and syngas can be used in power generation, heating, industrial boilers, cement kilns, and as soil amendments or chemical feedstocks. These findings support the effective use of LDPE as a plastic waste for bioenergy precursor production.

Keywords: Catalytic pyrolysis; Machine learning regression; Process optimization; Feature selection; Waste to energy (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225002907
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:317:y:2025:i:c:s0360544225002907

DOI: 10.1016/j.energy.2025.134648

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().