关键词: 毛细管电泳, 遗传算法, 功效函数, 对映体分离, 多指标优化

Abstract: A methodology based on the combination of desirability function approach and genetic algorithm was applied to the multicriteria-decision-making simultaneous optimization of epinephrine enantiomers by capillary electrophoresis. Uniform design, an experimental design method, was employed to examine how the factors such as electrolyte concentration, pH, concentration of chiral selector and voltage affected resolution, analysis time and peak height. In order to achieve the simultaneous optimization of the three goals for efficient, fast and highly sensitive separation, the Derringer desirability functions were tested. The real-value encoded genetic algorithm was used to acheive the search for maximizing desirability and several optimal separation conditions were obtained. Under the optimized conditions, the values of total desirability functions were increased by 10.4% in comparison to those under the optimal condition defined by the uniform design experiment.

Key words: Capillary electrophoresis, Genetic algorithm, Desirability function, Enantiomer separation, Multicritiria optimization