Optimization of Water Network Integrated with Process Models

CATEGORY: PIPING NETWORK
Chemical Engineering Transactions, Vol. 29, 2012
Optimization of Water Network Integrated with Process Models
Chun Deng*, Xiao Feng
State Key Laboratory of Heavy Oil Processing, College of Chemical Engineering, China University of Petroleum, 18 Fuxue Rd., Changping, 102249, Beijing, China
chundeng@cup.edu.cn
Abstract
Authors present a novel approach for the synthesis of a water network incorporated with process models. The process models are utilized to relate the process output (typically defined as internal water source) with process input (i.e. water sink). A generalized water network superstructure is developed to embed all possible process units and all the connections among resources, interceptors, process units and wastes.
The problem is formulated as four optimization problems (minimum freshwater flow rate, intercepted flow rate, intercepted mass load and number of connections) and the four models are solved in sequence to locate the targets. A case study is presented to illustrate the applicability and effectiveness of the proposed approach.
Introduction
In general, water network synthesis may be classified into two main categories, i.e. Fixed Contaminant load (FC) and Fixed Flow rate (FF) problems. In the former, water-using processes (e.g., washing, scrubbing, and extraction) are characterized by mass transfer operations where a fixed amount of contaminant is transferred from contaminant-rich stream to water, which acts as a mass separating agent. In contrast, water-using processes (e.g., boilers, cooling towers, reactors) are characterized as water sinks/sources that consume/generate a fix amount of water in the FF problems. Hence the primary concern of this latter problem is the water flow rate. As pointed out by Foo (Foo, 2009), the limiting water data for FC problem and FF problem are interchangeable. However, no matter the water-using processes belong to FC type or FF types, the characteristic of water-using processes could be specified via the appropriate process models, which associate the outlet process streams with the inlet process streams. Mathematical programming approaches have been considered as powerful tools for the synthesis of water conservation network. However, the water-using processes in previous models are typically classified into FC or FF types. In fact, the FC or FF types of water-using processes could be generally replaced
by process models.
In this paper, authors introduce a novel approach for the synthesis of water network incorporated with process models. The embedded process models are used to bridge the effluent of the water-using process (typically defined as internal water source) with the feed of the process (e.g. water sink). A generalized water network superstructure is developed to embed all possible process units and all the connections among resources, interceptors, process units and wastes. The proposed model is formulated as a sequential optimization problem. The applicability and effectiveness of the proposed approach is illustrated via solving a case study.
Free Full Text Source: http://www.aidic.it/cet/12/29/211.pdf