【作者】 郑雪松；

【导师】 冯霄；

【作者基本信息】 西安交通大学 ， 化学工程， 2005， 硕士

【摘要】 水系统集成技术,是节约过程工业新鲜水用量与减少废水排放量的最有效技术之一,过去二十年来有大量的相关技术被开发,其中主要包括了常规水回用网络优化技术和中间水道技术。 然而,对于常规水回用网络优化技术,人们主要关注于如何减少系统的新鲜水用量,而忽略了网络复杂性和系统的控制柔性;对于中间水道技术,现有的设计方法基于经验规则,无法保证最终设计方案的最优性。另外,目前考虑多污染物用水系统的能量特性,对用水系统与能量系统进行综合优化的研究也不多见。 本文针对不同用水网络结构进行了研究,主要内容如下: 针对常规水回用网络,在现有的用水网络超结构数学模型基础上进行了改进,建立了能够优化水网络结构的数学模型,从而使得最终的用水网络不仅能够达到较少的新鲜水用量,而且具有较少的物流连接数。这样的水网络更有利于网络铺设,并可降低投资费用。 针对具有中间水道的水网络,提出了通用优化设计方法。设计中先规定中间水道的最大级数,建立模型求解最小新鲜水用量。然后确定最大允许新鲜水用量,建立模型求解此条件下的最小中间水道级数。最后求解此中间水道级数下需要消耗的最小新鲜水量。最终可以得到新鲜水消耗量较小,网络结构比较简单,系统操作性较强的具有中间水道的用水网络。 提出了混合结构的水网络。混合网络为既存在中间水道又存在用水单元之间直接相连的水网络。建立了该网络的超结构,利用数学规划的方法对其进行优化设计。设计时首先计算用水系统在常规水回用网络中的控制数,参照此结果规定混合结构水网络的控制数。然后建立数学模型对新鲜水用量进行优化。最后建立数学模型对网络结构进行优化,减少中间水道的级数。混合结构的水网络可以兼顾新鲜水量、网络结构复杂性及用水系统控制柔性,综合了常规水回用网络新鲜水用量少,中间水道水网络结构简单的优点。 针对考虑热集成的用水网络,在用水系统超结构模型及换热网络超结构模型的基础上,建立了多目标数学规划模型。为了求解此多目标规划,提出了三种适用于不同特点系统的求解策略,分别以新鲜水量为主目标、以公用工程用量为主目标和以系统经济性为目标。对应于每种求解策略,多目标规划被转化为一个或一系列单目标规划进行求解。本模型既适用于单组分用水系统,又适用于多组分用水系统。换热物流可考虑用水系统以外的流股,使整个换热系统更多还原

【Abstract】 Water system integration is one of the most efficient technologies for saving freshwater and reducing wastewater. A lot of corresponding strategies have been reported over the past two decades, including the water pinch technology and the internal water main technology. But up to now, most efforts focused on how to minimize the freshwater consumption for the water-using network with traditional structure. The network complexity and the system controllability were overlooked in most research. As to the water-using network with internal water mains, the design methodology is based on algorithmic rules and cannot guarantee the optimum of the final solution. In addition, how to take account of heat integration in the water-using network design is not reported for multi-contaminant water-using systems. In this paper, research covers water-using networks with different structures as following: For the water-using networks with traditional structure, the mathematical model is improved by adding another programming to optimize the network structure. With this procedure, the final design not only can attain the minimum or acceptable freshwater consumption, but also can feature a comparatively simple structure. Due to the less number of connections between water-using processes, the final network is more convenient to construct and need less expense. For the water-using networks with internal water mains, a general design methodology is proposed based on the established superstructure of the network. Firstly, the maximum number of internal water mains is specified to define the scale of the water-using system, and a mathematical model is established to minimize the freshwater consumption of the system. Secondly, with the freshwater consumption limit, another model is established to obtain the number of necessary internal water mains. The last model can be established optionally to obtain the minimum freshwater consumption with specified number of internal water mains. The final design features a less freshwater consumption compared with the design without water reuse. It also has a simpler structure than the optimal water-using network with traditional structure. This contributes to its high controllability. The water-using network with mixed structure is proposed in this paper to balance the freshwater consumption and the network complexity. In such a network, internal water mains as well as the connections between processes can both exist in the final design. Mathematical models are established based on the superstructure to optimize the system. Firstly, the minimum number of potential control for traditional network is calculated as the reference. Then the minimum freshwater consumption under specified number of potential control can be obtained. The last programming is introduced to simplify the network structure. Water-using networks with mixed structure can take account of the freshwater consumption, the network complexity and the system flexibility simultaneously and integrate the advantages of the traditional structure and the internal water main structure. For the heat efficient water-using network, a mathematical model with multiple objectives is built to attain the minimum fresh water consumption and the minimum utilities simultaneously. Three solving strategies are proposed to convert it into sequential models with single objective. The first strategy is utilized in systems dominated by freshwater consumption, and the main goal is to reduce the freshwater consumption to the minimum. The second strategy is applied to systems dominated by utilities, and the main goal is to optimize the utility duty to the minimum. In the third strategy, economical concerns are taken into account to balance the importance of freshwater and utilities. This method can be applied to a water using system with multiple contaminants. In heat integration, not only the process-to-process streams in the water using system, but also the streams out of the water using system can be taken into account. In the end, the design method for the更多还原