Abstract: A simulated annealing based approach was developed for the synthesis of large-scale heat integrated distillation sequences.The synthesis problem was formulated as a mixed integer nonlinear programming (MINLP) problem into which a more efficient representation of network structure and a more rigorous cost model were introduced. With the consideration of the characteristics of the studied problem,the simulated annealing algorithm was improved so that it could handle structural and parametric optimization simultaneously within the same annealing scheme, and could attain reasonable trade-off between computational accuracy and speed. The proposed approach was shown to be effective by the solution of larger-scale examples (the number of products≥10).

摘要： 提出一个用于多组分热集成精馏系统综合问题的改进模拟退火算法.将优化问题表示为混合整数非线性规划（MINLP）问题，将算法设计同研究问题的特征相结合，运用本文方法可在同一退火策略下实现流程结构和操作参数的同步优化，并能达到计算精度和计算速度的更好均衡.通过对大规模算例（产品数≥10）的计算表明，该方法对求解此类问题非常有效.