a hybrid model of multi-objective differential evolution algorithm and various decision-making methods to optimize the batch abe fermentation process

In recent years, biofuels have attracted considerable attention as renewable and clean source of energy and have been playing the role of suitable alternatives to fossil fuels. one of the most attractive types of biofuels is acetone-butanol-ethanol (abe) which is produced in a batch fermentation process by the anaerobic bacterium clostridium acetobutylicum and sugar-based substrate as feedstock. in this paper, optimization of this process was carried out according to a bi-objective function. a hybrid model of multi-objective differential evolution (mode) algorithm and distinguished decision-making methods, namely linear programming technique for multidimensional analysis of preference (linmap), technique for order of preference by similarity to ideal solution (topsis) and shannon’s entropy were applied to find the final optimal operating point. the initial concentration of substrate and the final operating time of the process were selected as decision variables to maximize two main objectives in terms of solvents yield and productivity. a pareto optimal set presents a wide range of optimal operating points and a proper operating condition can be selected based on the necessities of the applicant. the best optimal point obtained by topsis, according to the lowest value of deviation index, was also compared with the results of economic-based optimization.