Optimization of Ethanol Fermentation using a Hybrid Model of Support Vector Machine and Genetic Algorithm

Abstract

Bioethanol is one of the most sustainable biofuels used in various applications throughout the world. This study aimed to optimize ethanol production from glucose fermentation using artificial intelligence techniques. A support vector machine (SVM) model was developed to predict ethanol production based on process simulation data. The predictions of the SVM model were assessed through multiple statistical measures, including the coefficient of determination (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ), root mean square error (RMSE), and maximum absolute error (MAE) for the training and testing datasets for various types of the model. The predictions of the examined SVM models scored a maximum R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.99 and 0.95 for the training and testing datasets, respectively. The best performing SVM model was coupled with a genetic algorithm (GA) to optimize the fermentation process parameters towards achieving the maximum yield of ethanol production. The SVM-GA results showed that a maximum ethanol production of 39,598 kg/hour can be achieved under the optimized conditions of 0.9 mass fraction, 100 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">o</sup> C temperature, and 10 bar pressure.