摘要
将Mod．UNIFAC基团贡献法与改进的遗传算法相结合,进行萃取精馏萃取剂的分子设计。针对遗传算法搜索效率低和不易得到全局最优解等问题,对遗传算法的编码方案和遗传操作算子中的选择和变异算子进行了改进,有效避免了该方法的自身缺陷。将其应用于乙醇-乙酸乙酯体系,设计得到了三甲苯、丙三醇、氯苯等高效萃取剂,将相对挥发度的预测值与文献值比较,平均偏差小于8%。说明了所采用的方法是准确可靠的。
A method based on improved genetic algorithm for the computer-aided molecular design for extractive distillation was presented. In order to simplify the synthetical molecules＇ forms and link easily to the data base for calculating molecules＇ physicochemical properties, a new encoding scheme was established, and it greatly enhanced performing efficiency of the algorithm. Cross generatent elist selection, dislocation crossover operators were developed to avoid the disadvantages of genetic algorithm with converging early and losing easily globally optimal results. Because of the improvement of genetic algorithm, the molecules＇ output structures were stabile and forms were diversified. The methodology and design process for the molecular design had been implemented by a computer program. The accuracy and reliability of the methods were verified by designing solvents for the extractive distillations processes of ethanol-ethyl acetate system, Trimethylbenzene as the industrially solvent and propanetriol, chlorobenzene, and so on, with higher relative volatilities were involved in the results. And the prediction precision of relative volatility of this work was better than the literature reported.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2006年第6期523-526,共4页
Computers and Applied Chemistry
关键词
遗传算法
基团贡献
分子设计
萃取剂
genetic algorithm, group contribution, molecular design, extractant
