Integrating Nanomaterial and High-Performance Fuzzy-Based Machine Learning Approach for Green Energy Conversion

Abstract

Biomass is a renewable and sustainable green energy material. It is made up of lignin, cellulose, and hemicellulose with considerable amount of water, extractives, and inorganic chemical compounds. The use of biomass materials and other biogenic wastes for energy recovery represents an eco-friendly way. Biomass material selection is one of the most significant aspects for any energy conversion process, and it is a common outsourcing problem that includes material preparation, reactor performance, economic assessment, and calorific value of the products. Fuzzy systems can be quite useful in high-performance computing during the selection of biomass materials. In each engineering process, material selection is a crucial step since each material is having its own set of characteristics. This study presents the application of type-1 fuzzy set for the selection of suitable biomass material for yielding maximum bio-oil. This study focuses on seven locally available materials such as rice straw (M-1), sunflower shell (M-2), hardwood (M-3), wheat straw (M-4), sugarcane bagasse (M-5), corn cop (M-6), and palm shell (M-7). The study evaluated seven important properties of the materials such as lignin (P-1), cellulose (P-2), hemicellulose (P-3), volatile matter (P-4), fixed carbon (P-5), moisture content (P-6), and ash content (P-7). The findings demonstrated that sugarcane bagasse (M-5) is the best option for maximum bio-oil yield. Furthermore, the potential of nanoscale catalysts in improving the yield of bio-oil through real-time experiments was studied. The findings of this work add to our understanding of the application of fuzzy-based systems for energy applications.