Document Type : Research Paper
Abstract
The increasing demand for alternative renewable energy sources has driven research into bioethanol production as a sustainable solution. This study investigates the production of bioethanol from cassava waste slurry using Saccharomyces cerevisiae. Process optimization and predictive modeling were conducted using Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS). The research involved experimental fermentation, data collection, and computational modeling to analyze the effects of temperature, pH, and sugar concentration on ethanol yield. A total of 19 experimental runs were conducted, and the models were evaluated based on statistical indices, including mean square error (MSE) and coefficient of determination (R²). The results demonstrated that ANFIS exhibited superior predictive accuracy, with an R² value of 0.9999 and a minimal error of 0.0000013607, outperforming ANN. The optimal conditions for bioethanol production were identified as 32.5 °C, 0.238 mol/dm³ sugar concentration, and pH 5.25, yielding 30% ethanol. Fourier Transform Infrared (FTIR) spectroscopy confirmed the presence of ethanol, validating the effectiveness of the fermentation process. The findings highlight the potential of cassava waste as a viable feedstock for sustainable bioethanol production and emphasize the advantage of soft computing techniques in optimizing bioethanol yield. These results contribute valuable insights into biofuel research, paving the way for efficient, cost-effective, and environmentally friendly bioethanol production processes.
