针对苯乙烯工艺流程的设计及控制复杂性,使用Kriging代理模型方法,对由Aspenplus构建的苯乙烯工艺流程模拟系统进行建模,并且使用分布估计算法对目标函数进行优化。结果表明,基于Kriging模型的分布估计算法,在保证较高精度要求的前提下,明显减少了计算时间,有效地化解了精度和效率之间的矛盾,具有理论意义及实用价值。
Aiming at the design and control complexity of the styrene manufacturing process, Kriging surrogate model is used to model the styrene production process simulation system constructed by Aspen plus. Moreover, the distributed estimation algorithm is adopted to optimize the objective function. Results show that the proposed algorithm can not only ensure high-precision requirements, but also significantly reduce its calculation time and effectively cope with the contradiction between precision and efficiency.