Optimization of Operational Parameters for Bio‑Oil Production from Municipal Solid Waste via Microwave Pyrolysis
2423-5547
Abstract
The efficiency of bio-oil production from municipal solid waste was investigated using microwave-assisted pyrolysis. The effects of temperature, pyrolysis time, and microwave power on bio-oil yield were evaluated using Response Surface Methodology (RSM), and optimal operating conditions were determined. In parallel, seven machine learning algorithms were applied to predict bio-oil yield, among which the Gaussian Process Regression model showed the highest performance, with an R² of 0.9714. Temperature was identified as the most influential factor in the prediction. The produced bio-oil was characterized in terms of its physical properties and chemical composition. The optimal process conditions were found to be 329.82 °C, 28.47 min, and 447.38 W. Temperature showed an initial increasing and then decreasing effect on yield, with higher temperatures favoring lighter hydrocarbons, while pyrolysis time exhibited a consistently positive effect on bio-oil production. These results confirm the effectiveness of microwave pyrolysis for converting municipal waste into bio-oil and highlight the value of combining RSM and machine learning for process optimization.