Optimization of erosion performance of biomass and pet waste based composites using artificial neural network

Abstract

Abstract The determination of the potentiality of renewable energy resources holds significant importance, with biomass emerging as a crucial alternative for both energy and material needs. Consequently, predicting the mechanical properties of these resources has become a focal point. This study focuses on the analysis of fundamental products resulting from the pyrolysis process, specifically char, extracted from Polyethylene terephthalate (PET) Char, Cashew biochar, and Sugarcane biochar and examining erosion performance of polyester composites. The polyester composites subjected to erosion tests to determine their wear resistance at various impact angles. Among the studied composites, those including cashew biochar shown enhanced erosion resistance, with the least erosive wear at a 60° impact angle. The investigation aims at optimizing the erosion performance of these biomass-based composites using an Artificial Neural Network (ANN) model. The ANN was trained to predict erosive wear behavior using input factors as biochar type, filler content, and impact angle. The model effectively found ideal conditions for decreasing wear, demonstrating the potential of Cashew biochar-filled composites for applications needing high erosion resistance. This work sheds light on the successful usage of biochar fillers in improving the durability of polyester composites, presenting a sustainable alternative for materials engineering.

Keywords

• Agricultural waste
• Bio char
• Polymer composites
• Erosive wear
• ANN