论文部分内容阅读
提出采用遗传规划求解精馏系统综合问题,制定了一套独特的编码方案和求解策略,可以利用树状结构代码直观地表达各类精馏流程,而且算法能够灵活表达任意可行的塔间热集成结构。同时,结合分离工程领域知识指导代码生成和进化操作,以保证代码的有效性,从而明显提高了算法的进化速度。在不必给定精馏系统超结构的前提下,算法成功实现了可行域的自动搜索,从中快速求解出最优精馏结构。计算表明,该算法能够快速准确地求解大规模精馏系统最优综合问题,为工业生产中精馏流程的最佳设计和优化操作提供准确依据。
Proposed the use of genetic programming to solve the comprehensive distillation system, developed a unique encoding scheme and solution strategy, you can use the tree structure code intuitive expression of various types of distillation process, and the algorithm can flexibly express any feasible hot tower integration structure. At the same time, combined with the separation of engineering knowledge to guide the code generation and evolutionary operations in order to ensure the validity of the code, thus significantly improving the evolutionary speed of the algorithm. Under the premise of not having to specify the superstructure of the distillation system, the algorithm has successfully realized the automatic search of the feasible domain, from which the optimal distillation structure can be solved quickly. The calculation shows that this algorithm can solve the optimal synthesis problem of large-scale rectification system quickly and accurately, and provide an accurate basis for the optimal design and optimization of the distillation process in industrial production.