Machine learning utilization on air gasification of polyethylene terephthalate waste

Studies in the field of machine learning utilization on air gasification of polyethylene terephthalate (PET) waste are of utmost importance and can contribute to solving the environmental challenges associated with PET waste while also promoting the development of advanced technologies in the field of waste management and renewable energy. The primary objective of this study is to focus on the gasification process of PET waste through the utilization of machine learning algorithms. The aim is also to assess how well these algorithms can predict and evaluate the gasification performance of PET waste. To achieve this, a model for air gasification of PET waste is created, and machine learning algorithms are developed and evaluated based on their performance. The results suggest that the H2/CO model has a high accuracy, as indicated by its R-sq value of 91.86 %. It is important to highlight that models developed for the lower heating values and cold gas efficiency show excellent accuracy, with R-sq values of 99.84 %. The high predicted R-sq values of models (higher than 90 % for the H2/CO model and higher than 99 % for models developed for the lower heating values and cold gas efficiency) indicate that these models excel in predicting future observations with great accuracy.