【作者】 李丽娟；

【导师】 董宏光；

【作者基本信息】 大连理工大学 ， 化学工程， 2011， 硕士

【摘要】 质量交换网络,包括吸收、解吸,萃取与离子交换等生产过程,是过程系统工程中一个重要的研究分支。典型的质量交换网络,即对于含有可回收或者有害组分的富流股,通过与能够接受这些组分的过程贫流股或者外加质量分离剂进行逆流操作实现质量传递。质量交换网络的综合课题的研究,不仅可以提高产品纯度和物料利用率从而来增加企业的经济效益,而且还可以减少污染物的排放来达到良好的环保效益。本研究旨在构建起一套系统化的研究方法来同时优化连续过程以及间歇过程中的质量交换网络的操作费用和设备投资费用。总体而言,本研究基于状态空间超级结构拓扑出系统所有可行的结构,然后在该超级结构基础之上以年度化总费用为目标函数建立起质量交换网络的混合整数非线性规划模型。此外,针对所建立的非凸的混合整数非线性规划模型还提出了随机型算法与确定型算法相结合的求解策略,同时增强了求解质量和效率。研究主要内容为：(1)首先从热力学角度分析了连续质量交换网络综合的特点及研究难点,并利用状态空间超级结构的基本框架深入探讨了流股在整个浓度区间的非等浓度混合以及非同源流股的混合分配机制。然后依托该结构建立起质量交换网络的混合整数非线性规划模型,来对系统网络的年度化总费用(TAC)进行优化求解,统筹权衡了操作费用和设备投资费用。(2)为了更好地挖掘过程流股循环回用的可能性,从而达到提高物料利用率并减少废物排放量的目的,本论文还对质量交换的初级网络与再生网络进行了同时优化。并且通过引入能量分离剂ESAs,本研究进一步研究了传统质量交换网络与利用ESAs的精馏系统的同时优化问题。通过与文献的比较,本论文所提出的优化策略均可有效地降低网络费用,同时减少新鲜物料的使用和废物的排放。(3)近年来,间歇过程由于特别适宜小批量、高附加值的精细化工产品的生产而受到越来越多的重视。间歇生产会产生一系列随着时间推进而不断变化的贫、富流股,它们之间可以通过物质回收来减少污染排放。间歇过程的质量交换网络的设计问题具有很大的挑战,比如贫、富流股的匹配不仅要受到流股的流量和浓度要求,并且还要遵循一定的时间序列的约束；而出现在不同时段的流股若要回用则需引入存储单元来实现物质的均衡。本论文所构建的状态空间超级结构不仅拓扑出系统完备的连接结构,还保证了在时间和浓度两维上同时优化存储单元所处位置。研究中,同时考虑了有／无存储单元的案例,对储罐在间歇网络综合中的作用进行了分析探讨。更多还原

【Abstract】 Mass exchange network (MEN), a system including absorption, stripping, adsorption, extraction, leaching and ion exchange, is of great significance as an important branch in Process Systems Engineering. In a typical problem of mass exchange network design, the valuable or undesirable components in rich process streams can be reduced in countercurrent direct-contacted mass exchange units by contacting with lean process streams or external mass separating agents (MSAs). Techniques of mass exchange network design can be used to improve the materials’purity and the utilization of valuable resources to satisfy the economic benefits, and also reduce the poisonous emissions for the purpose of pollution prevention.This paper is aimed at developing a systemic methodology for minimizing the operating costs and capital costs of the mass exchange networks for both batch and continuous processes in a simultaneous manner. Generally speaking, the State Space Superstructure is adopted to capture all possible network configurations, based on which better total annualized cost (TAC) can be easily generated with the corresponding mixed-integer nonlinear programming (MINLP) model. In addition, a hybrid optimization strategy incorporating both deterministic and stochastic components is developed for the resulting MINLP model to guarantee the solution quality and efficiency. The main contents addressed in this paper can be stated as follows:(1) The characteristics and difficulties of the study in mass exchange network design for continuous processes are first investigated from the views of thermodynamics. By introducing the State Space Superstructure, this paper investigates the mixing/splitting strategies of the streams with different compositions and sources. Furthermore, the mixed-integer nonlinear programming (MINLP) model is formulated based on the State Space Superstructure to optimize the total annual cost (TAC) of the MENs, so that a trade-off between operating cost and capital cost can be achieved.(2) In order to utilizing the materials more sufficiently and thus reducing the harmful emissions, simultaneous optimization of the preliminary network and regeneration networks in the mass exchange systems is investigated in this paper. Moreover, by introducing the Energy Separating Agents (ESAs), this paper further analyses the combined optimization problem of traditional mass exchange network and distillation system, which is corresponding to the use of ESAs. According to a serious of case studies, the proposed model formulation and solution strategy can not only decrease the total cost associated the MENs, but also reduce the consumption of fresh materials and the emission of wastes.(3) Batch productions have aroused ever-increasing awareness for the past decades, due to their incomparable advantages in producing small-amount and high-value-added products. These processes would generate many time-dependent rich or lean process streams with great potential of mass recovery and environmental protection. However, there are several challenges in batch MEN design, such as matches between process streams from different time sequences and the storage strategies, that make the optimization problem as a relatively rare-discussed topic compared with the continuous one. In this work, the State Space superstructure has been improved to capture all the potential structures, including stream splitting and mixing, matches between rich and lean streams, and also allows storage tanks to locate at any possible composition point and any time interval. Furthermore, synthesis problems of the batch mass exchange network with/without storage tanks are both taken into consideration, in order to analysis of the roles of storage strategies in network design for the batch processes.更多还原