Optimization of an eco-friendly municipal solid waste-to-multi-generation energy scheme integrated by MSW gasification and HSOFC: Regression analysis and machine learning study

Abstract

The utilization of regression models and machine learning algorithms in waste-to-energy systems has the potential to enhance the efficiency and effectiveness of these systems. The predictive capacity to ascertain energy production and optimize operational methodologies can contribute to increased efficacy and proficiency of waste-to-energy systems, thereby diminishing waste generation and fostering the generation of sustainable energy. The gasification technology for Municipal Solid Waste (MSW) provides a multitude of environmental advantages and holds significant importance in safeguarding the environment. This approach entails the use of sustainable and environmentally friendly practices in waste management, which effectively mitigates the release of greenhouse gases while also preserving finite natural resources. Considering the plethora of benefits it offers, it is anticipated that the utilization of MSW gasification as a waste management technique will experience a noticeable surge in popularity in the near future. This study presents a modeling approach for a multi-generation energy system that utilizes municipal solid waste (MSW) gasification technology integrated with a proton conducting solid oxide fuel cell (SOFC). Machine learning algorithms have been devised for the purposes of predicting and optimizing the performance of systems. The heating generation, power generation, electrical efficiency, exergy efficiency, and emission levels of the system have been accurately predicted, exhibiting a substantial level of precision as indicated by R2 values predominantly exceeding 94%. The optimization procedure reveals that the optimal solutions entail a heating generation rate of 854.2 g/s, power generation of 323.6 kW, an electrical efficiency of 40.3%, an exergy efficiency of 36.1%, and an emission rate of 893.5 g/kWs