【作者】 王达；

【导师】 郑世清； 岳金彩；

【作者基本信息】 青岛科技大学 ， 化学工程， 2005， 硕士

【摘要】 过程系统综合是指按照规定的系统特性,寻求所需要的系统结构及其各子系统的性能,并使系统按规定的目标进行优化组合。它是过程系统工程学的核心内容,是过程系统设计的关键。随着近年来在过程集成设计中需要同时考虑环境性、经济性和可操作性等因素,化工过程的多目标优化综合成为一个重要的研究课题。多目标过程优化综合问题为多目标混合整数非线性规划模型的求解。多目标优化问题与单目标优化不同,多目标优化问题的特点是极少存在绝对最优解,而是存在一个非劣解集(Pareto解集),多目标优化技术的主要目的就是寻求Pareto解集中的一个或多个满意解。求解方法主要有数学规划法和多目标进化算法。以多目标遗传算法为代表的进化算法被认为特别适合求解此类问题。遗传算法大多用于单目标问题的优化,近十几年来将遗传算法应用到多目标优化的研究得到了很大的发展。 本文在充分研究了多目标过程优化综合方法的基础上,在化工通用模拟软件ECSS—化工之星平台上开发了一个多目标化工过程优化综合系统。本系统采用遗传算法。 本系统对精英保留的非劣排序遗传算法(Non-dominated Sorting Genetic Algorithm,NSGA-Ⅱ)进行了改进,针对化工过程的模型特点,对改进的精英保留的非劣排序遗传算法(NSGA-Ⅱ)在过程综合中的应用研究进行了讨论,并认为该算法应是求解此类问题的有效算法。 本文开发了具有多目标优化综合功能的模块,将该模块集成于化工模拟软件ECSS—化工之星,为过程优化综合提供了方便而可靠的平台,实现过程模拟软件的过程优化综合功能,满足用户对化工过程多目标优化综合的需要。更多还原

【Abstract】 The process system synthesis is refers that seek system structure and its various subsystems capability according to the system characteristic, and causes the system to carry on the optimized combination according to the stipulation goal. It is the process system engineering core content, is the key of process system design. Because in the process integration design needed to consider the environment, the efficiency and the maneuverability simultaneously, the chemical process multi-objectives optimization synthesizes become an important research topic. This question almost solves the model of Multi-objectives Mix-integer Nonlinear Programming (MOMINLP). The multi-objectives optimization is different from the simple objective optimize, there is few absolute optimal solution to multi-objectives problem, but has the set of non-dominated solutions which called Pareto-optimal front. The multi-objectives optimization technology main goal is seek one or manysatisfactory solutions in the Pareto-optimal front. The solution methods mainly have mathematics programming and the multi-objectives evolution algorithms. The Multi-objectives Genetic Algorithms (MOGA) as representative evolution algorithms was considered specially suit to solve this kind of questions. The genetic algorithms mostly use in the single objective questions optimization, in recent years the research that apply the Genetic Algorithms to the multi-objectives optimized has been developed.This article based on the fully studying in the multi-objective optimization method, has developed a multi-objective chemical process optimization synthesis system on the platform of the chemical industry general simulation software ECSS. This system uses the Genetic Algorithm.This system has made the improvement to Non-dominated Sorting Genetic Algorithms Ⅱ (NSGA-Ⅱ). In view of the characteristics of chemical process model, discuss applied research of the improved Non-dominated Sorting Genetic Algorithms II (NSGA-II) in the process synthesis. And considers that t the improved Non-dominated Sorting Genetic Algorithms is the effective algorithms to solve this kind of questions.This article developed the module which had the multi-objective optimization synthesis function. This module integer to the chemical industry simulation software ECSS, provided the convenience but reliable platform for the process optimization synthesis, realized process optimization synthesis function to process simulation software, satisfied the user’ s demands to the chemical process multi-objectives optimization synthesis.Example indicated that this system can be satisfied the更多还原