用于间歇化工过程最优设计的遗传算法

A GENETIC ALGORITHM FOR OPTIMAL DESIGN OF BATCH PROCESSES

间歇化工过程的最优设计问题是一类复杂且难以求解的组合优化问题。通过把这类问题分解为只包含离散变量的主导问题和只含连续变量的子问题，把遗传算法和线性规划法结合起来对其进行求解。并在算法中引入了一类新的算子，显著地提高了收敛概率。算例表明，该方法可以避免直接求解过程的复杂性和困难，并且具有很好的全局收敛性。

Optimal design of batch processes is a complex combinatorial optimization problem with difficulties in solution finding.In this paper,the original problem of optimal design is decomposed as a master problem and sub problems.An algorithm combining genetic algorithm (GA) and linear programming(LP) is applied to solve the optimal design problem.New operators are proposed to improve the possibility of convergence.Examples show that the algorithm can avoid the difficulty commonly encountered in direct solution procedures and give good global convergence.