A design methodology for multiple-contaminant water networks with single internal water main

https://doi.org/10.1016/S0098-1354(02)00177-1Get rights and content

Abstract

In this paper, the application of the water networks with internal mains is introduced. A design methodology for multiple-contaminant water networks with single internal water main is presented. A new concept of ‘water-saving factor’ is proposed. Emphasis is placed on the location of the first internal water main, which is related to the maximum water-saving potential. The paper is accompanied by several case studies to illustrate the methodology. According to these case studies, water networks with just one internal water main determined by the presented method can obviously reduce water consumption, approaching the minimum water consumption target.

Introduction

In recent years, water scarcity and stricter environmental regulations on industrial effluents underlie the growing emphasis on fresh water minimization in industry, which corresponds to wastewater minimization.

Being an efficient technology for saving freshwater and reducing wastewater, water system integration becomes the research focus as it can result in maximum water-saving effect. It regards the water-using system in a plant as a whole by considering how to allocate the quantity and quality of water in each water-using unit so as to maximize water-reusing and minimize wastewater discharge.

In 1980, Takama, Kuriyama, Shiroko and Umeda (1980) addressed an approach for optimal water allocation in a petroleum refinery based on a superstructure of all possible re-use and regeneration opportunities. Then in 1989, El-Halwagi and Manousiouthakis (1989) defined the composite curve to denote mass exchange operation, which was adapted from the methodology developed for heat exchanger networks by Linnhoff and Hindmarsh. Remarkable work was done by Wang and Smith (1994a) by introducing the important concepts of ‘water pinch’ and ‘limiting water profile’, which is further improved later (Doyle & Smith, 1997, Kuo & Smith, 1998, Wang & Smith, 1994a, Wang & Smith, 1994b, Wang & Smith, 1995). The graphical method can give the minimum fresh water requirement of the entire process in a direct way, but when multiple-contaminants are present, graphical methods require assumptions for ease of implementation, some of which may be difficult to justify. Huang, Chang, Ling and Chang (1999) presented a mathematical programming solution of the combined problem of water allocation and treatment by solving nonlinear problem (NLP). Recently, Bagajewicz and Savelski (Bagajewicz, Rivas & Savelski, 1999, Savelski & Bagajewicz, 1999a, Savelski & Bagajewicz, 1999b) showed how to reduce a nonlinear program to a linear program (LP), by inserting the maximum outlet concentration conditions. The mathematical programming methods are effective in optimizing large-scale systems but can be difficult to interpret, giving designers fewer insights compared with graphical methods. Works by others are also in the two kinds (Alva-Argaez, Vallianatos & Kokossis, 1999, Castro, Matos, Fernandes & Nunes, 1999, Olesen & Polley, 1997, Polley & Polley, 2000, Li & Yao, 2001).

Water networks mentioned above are designed using strategies similar to those used in the design of networks of heat exchangers, with pipes connecting the unit processes. When the entire plant involves only a few processes, the networks are fairly simple and water-savings at or near the maximum are achieved. However, for a large petrochemical or chemical complex, with many unit processes, the piping network becomes very complicated and hard to design and control. Moreover, the change of water flowrate or water quantity in one unit process will influence other unit processes. If using freshwater to adjust the change, the water-saving effect will decrease.

To solve the problem, Feng and Seider (2001) introduced a new structure and its design methodology for water networks. The new structure uses one or more internal water mains (or reservoirs) to simplify the network. These water mains simplify operation and the control of water quality, as well as the design strategy. However, their work involved only single contaminant. When dealing with the multiple-contaminant problems, the design methodology for single contaminant cannot be used anymore. In multiple-contaminant system, the key contaminant of each unit is different which is related to the water source to the unit. So the inlet and outlet flowrate allocation of each unit is difficult to specify because its freshwater consumption is close related to the choice of its sources. Due to the complexity, the design for water networks with single contaminants based on internal mains cannot be extended to that of multiple-contaminant systems. In this paper, the design methodology for multiple-contaminant systems is presented. Emphasis is placed on the location of the first internal water main. Generally, water networks having just one internal main determined by the presented method can obviously reduce water consumption, approaching the minimum value.

Section snippets

Network with internal water mains

All plants contain external mains including fresh water and wastewater mains.

The internal water main is a reservoir at a uniform concentration of contaminants. It receives water at contaminant concentrations less than or equal to its contaminant concentrations, and supplies water to unit processes at concentrations less than or equal to their contaminant concentrations. Like the fresh water main or the steam pipe, the internal water mains are connected to many unit processes.

Positioning

Design procedure of water networks with single internal water main

For easy to understand, the design procedure is illustrated with an example. The limiting flowrate and terminal concentrations in example 1 are shown in Table 1. The minimum freshwater target is 139.3 t/h. This example involves one internal water main whose contaminant concentration is specified. Specifying the location of the internal water main will be discussed in the next section.

Determining the concentration of the internal water main

Appropriately positioning the internal water main can simplify the network and obviously reduce the freshwater consumption. Feng and Seider (2001) introduced a design methodology for single contaminant system, which set one or two mains according to the location of pinch and get good water-saving effect, but for multiple-contaminant system, the concentration of internal water main cannot be specified from pinch. Here an example is used to give the method of determine the concentration of the

Conclusions

Water networks with internal water mains are shown to provide simpler water networks, which is easy to design, operate and control, for plants involving many unit processes, such as those in petrochemical and chemical complexes.

Generally, water networks with just one internal water main determined by the presented method can sharply reduce water consumption, approaching the minimum target.

Appropriately positioning the internal water main can give a simple network and obviously reduce the

Acknowledgements

Financial support from the National Nature Science Foundation of China (Project 20176045), the Major State Basic Research Development Program of China (Grant No. G20000263), and China Petroleum & Chemical Corporation (project J301005) is thankfully acknowledgement.

References (17)

There are more references available in the full text version of this article.

Cited by (79)

  • Total Site Material Recycling Network Design and Headers Targeting Framework with Minimal Cross-Plant Source Transfer

    2021, Computers and Chemical Engineering
    Citation Excerpt :

    Feng and Seider (2001) introduced the plant-level centralised headers/main concept for the water recycle or reuse network design with concentration identified certain heuristic. Wang et al. (2003) extended it for multiple contaminants, and Cao et al. (2004) extended the concept accounting for the regenerated water main. Using a mathematical approach, Chen et al. (2010) developed a comprehensive mixed-integer non-linear programming (MINLP) method for Total Site Integration with centralised and decentralised water mains.

  • Optimisation, retrofitting, and variable condition analysis of cooling water networks with improved intermediate main configurations

    2021, Applied Thermal Engineering
    Citation Excerpt :

    Few researchers conducted in-depth studies on the change in load conditions that focussed on both the flexibility and economic performance of the cooling water network. An approach using an intermediate water main [31,32] was proposed, in which water reuse was implemented using an intermediate water main. In this configuration, water consumption was reduced suitably compared with the parallel configuration.

View all citing articles on Scopus
View full text