An artificial intelligence study on energy, exergy, and environmental aspects of upcycling face mask waste to a hydrogen-rich syngas through a thermal conversion process

Studies in the field of gasification process for face mask wastes play a significant role in addressing the environmental issues associated with these wastes and harnessing their potential as an important energy source. On the other hand, machine learning techniques offer a promising approach to predict the environmental, energy, and exergy performances of the gasification process. This study addressed the utilization of machine learning techniques on the gasification process of face mask wastes. Various polynomial regression machine learning algorithms are developed for syngas lower heating value, exergy and cold gas efficiencies, and emission level and their performances are checked. Remarkable precision of machine learning algorithms in forecasting the response variables is evidenced by the impressive outcomes. In fact, the majority of cases yield R-sq values surpassing 98%, thus highlighting their effectiveness. Specifically, when it comes to exergy and cold gas efficiencies, and emission, the machine learning algorithms achieve exceptional R-sq values of 99.85%, 98.93%, and 99.33%, respectively. These exceptional results confirm the credibility and accuracy of these algorithms in providing precise predictions. Leveraging the power of artificial intelligence not only contributes to the development of more sustainable energy systems but also helps in making informed decisions regarding process design and operation.