摘要
应用遗传算法估算溶液热力学模型参数,并对标准遗传算法中的变异策略和竞争方式作了适当的改进,得到改进的遗传算法。举DMF(二甲基甲酰胺)+water体系的溶液热力学模型参数的估算为例,并与POWELL法比较。计算结果表明,遗传算法比POWELL法具有更强的寻优能力,而本文所提出的改进的遗传算法比标准遗传算法的寻优速度明显较快,对解决溶液热力学模型这类复杂的非线性函数的参数估算问题,本文所提出的改进的遗传算法是一种较好的寻优算法。
Genetic algorithm was used to estimate the parameters of thermodynamic model of solution, and a modified genetic algorithm was proposed by modifying the strategy of mutation and the mode of competition. DMF + water VLE system was taken as an example, and the results were compared with those of POWELL method. It is found that the optimization ability of genetic algorithm is much more powerful than that of POWELL method and optimization speed of the proposed modified genetic algorithm is much greater than that of standard genetic algorithm . The modified genetic algorithm is valid for parameter estimation of thermodynamic model of solution.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2003年第1期171-174,共4页
Computers and Applied Chemistry
基金
福建省自然科学资助项目(B9810011)
关键词
改进遗传算法
溶液热力学模型
参数估算
modified genetic algorithm
thermodynamic model of solution
parameter estimation
