Format And Type Fonts CHEMICAL ENGINEERING TRANSACTIONS VOL. 45, 2015 A publication of The Italian Association of Chemical Engineering www.aidic.it/cet Guest Editors: Petar Sabev Varbanov, Jiří Jaromír Klemeš, Sharifah Rafidah Wan Alwi, Jun Yow Yong, Xia Liu Copyright © 2015, AIDIC Servizi S.r.l., ISBN 978-88-95608-36-5; ISSN 2283-9216 DOI: 10.3303/CET1545024 Please cite this article as: Jia X., Li Z., Wang F., Chen C.-L., Foo D.C.Y., 2015, Inter-plant water integration with considerations of water supply constraint and differential water price, Chemical Engineering Transactions, 45, 139-144 DOI:10.3303/CET1545024 139 Inter-plant Water Integration with Considerations of Water Supply Constraint and Differential Water Price Xiaoping Jia* ,a , Zhiwei Li a , Fang Wang a , Cheng-Liang Chen b , Dominic C.Y. Foo c a School of Environment and Safety Engineering, Qingdao University of Science & Technology, Qingdao 266042, China b Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan c Centre of Excellence for Green Technologies/Department of Chemical and Environmental Engineering, University of Nottingham Malaysia, Broga Road, 43500, Semenyih, Selangor, Malaysia jiaxp@qust.edu.cn The recent increase in awareness of industrial ecology has inspired research in inter-plant water integration. Industrial water conservation and wastewater reduction are becoming increasingly important for sustainable water resource development in industrial parks. Water supply and pricing structures provide incentives to increase water recovery ratio. This work proposes a superstructural-based mathematical optimization method for total water networks with constrained freshwater supply. This work first explores the implications of differential price, minimum freshwater flowrate, and the number of regeneration units for different constrained water supplies. Considering minimum water-using cost as the optimization objective, the minimum freshwater flowrate and its corresponding total water network structural are determined. 1. Introduction Water resource scarcity and increasing water prices in the process industries have spurred research efforts focusing on the reduction of freshwater consumption and treatment costs through water reuse, recycling, and regeneration. These problems have been addressed as part of a synthesis of intra-plant and inter-plant water integration from a process system engineering perspective. Many systematic design methods for water conservation based on process integration methodologies have been emerged. Such methodologies reported for intra-plant water networks have been summarized in the review articles by Bagajewicz (2000) for fixed flow problems, Foo (2009) for fixed flow rate problems, as well as the literature annotations by Jeżowski (2010). The recent increase in awareness of industrial ecology has inspired research in inter-plant water integration (IPWI). Olesen and Polley (1996) first addressed practical considerations in IPWI by analyzing the influence of geographical location and piping costs using the pinch-based load table technique. Two different schemes of IPWI is proposed, i.e., unassisted Integration (Chew et al., 2010a) and assisted integration (Chew et al., 2010b). Rubio-Castro et al. (2013) presented a new global optimization method for IPWI problems based on properties and formulated a MINLP model that involves all possible options of interest. Tan et al. (2011) developed a fuzzy bi-level programming model to determine optimal inter-plant water integration networks in eco-industrial parks. Lee et al. (2013) presented a mathematical model for the synthesis of water networks involving mixed batch and continuous units. Chen et al. (2010) proposed IPWI scheme with central and decentralized water mains. In addition, Chew et al. (2009) employed a game-based approach to analyse the interaction of participating companies in an eco-industrial park. Sueviriyapan et al. (2014) develop a generic model-based synthesis process for a water/wastewater treatment network design problem to identify the best wastewater treatment processes with a minimum total annualized cost. Pintarič et al. (2014) performed water networks syntheses based on MINLP model by using various economic objectives, i.e. total annual cost, the net present value, and the annual profit. The techniques aforementioned may be used to find the minimum freshwater consumption for IPWI problems. However, the synthesis of water networks is a complex task, as there are many parameters that 140 influence water reuse/recycle problems. These include the total benefit/cost, constrained freshwater supply, and water price. Jia et al. (2010) proposed a mathematical optimization model for total water networks with constrained water supply and differential water pricing. The aim of this work is to explore the relationships between differential price, minimum freshwater flowrate, freshwater supply, and the number of regeneration units. 2. Methodology 2.1 Problem statement and superstructural models Given a total water network, it consists of a freshwater supplier, a set I of water-using plants, a set T of wastewater treatment plants, and a water regeneration plant with R water regeneration units. Assume that some basic data are known, including the limiting flowrate, the limiting inlet and outlet concentrations, and the types of contaminant. Additionally, all cost elements are known, such as regeneration water price, regeneration cost, the restricted water flowrate, and freshwater price. The superstructures for water-using plant, water regeneration plant, and wastewater plant are shown as Figure 1. In Figure 1(a), there are three water inlet streams: the freshwater or available water, the water stream from a regeneration unit with low concentrate, and the reuse water from other water-using plants. These water streams mix in a mixing unit and then enter the water-using plant. During the process, water as a mass separating medium aims to remove the contaminant load in the water-using plant. The outlet streams are divided into three sub-streams by a splitter. These sub-streams enter another water-using plant, the regeneration plant, and the wastewater plants. The mass balance of the mixer is shown in Eq(1) and the constraints of the mixer are shown in Eq(2) – Eq(5). The mass balance of contaminant unit i is shown in Eq(6) and the constraints are shown in Eq(7). The mass balance of the separator for water-using plants are given as shown in Eq(9) – Eq(11). Water-using plant i From water supplier From another water-using plant From water regeneration plant To another water-using plant To wastewater treatment plant To water regeneration plant i  i r , , i i i i i out e F Z C    , , ri ri r out e F Z C wi wi w F Z C , , ir ir i out e F Z C , ,it it i out e F Z C , , ii i out e ii F Z C   M S , , out i i out e F C, , in i i in e F C , load i e M t (a) Water regeneration plant r From water supply plant From another water-using plant i From water regeneration plant To another water-using plant To treatment plant To water regeneration plant i r , , ir ir i out e F Z C , , r r r r r out e F Z C    ,wr wr w e F Z C , , rr rr r out e F Z C   , ,rt rt r out e F Z C , , ri ri r out e F Z C   M S , , in r r in e F C , , out r r out e F C , load r e M r t (b) Wastewater treatment plant t From water supply plant From another water-using plant i From water regeneration plant r , , it it i out e F Z C , . rt r out e rt F Z C ,wt wt w e F Z C M , , out t t out e F C , , in t t in e F C (c) Figure 1 The superstructure for (a) water-using plants; (b) regeneration plants; (c) wastewater treatment plants (M: mixer; S: separator) Mass balance of the mixer: IiFFFF Ir ri Ww wi Ii ii in i    , '' '' (1) 141 IiCFCFCFCF Ir eoutrri Ww ewwi Ii eoutiiieini in i    , ,,,,,,, '' '''' (2) IiiZFFZF iiiiiiiiii  '' ,,'''''''''' (3) WwIiZFFZF wiwiwiwiwi  ,, (4) RrIiZFFZF ririririri  ,, (5) Where in i F is the outlet flowrate of the mixer supplied for water-using plant i , '' ,i i F is the flowrate from water-using plant ''i to plant i . wi F and ri F present the flowrate from freshwater and regeneration unit, respectively. , ,i in e C denotes the concentration of contaminant e at the inlet of the plant i . '',out,i e C , ,w e C and , ,r out e C represent the concentration contaminant e from water-using plant '' i ’, freshwater and regeneration unit, respectively. ''i i Z , wi Z , riZ is a binary variable denoting the existence of the link between the water source and water sink. ''i i F , wr F , riF is the lower bounds while ''i iF , wiF , riF is the upper bounds. Mass balance of contaminant of unit i : eini in ieouti out i load ei CFCFM ,,,,,  (6) EeRiCCCC eoutieoutieinieini  ,,; max ,,,, max ,,,, (7) Where , load i e M is the removal mass load of contaminant e of plant i . out i F and , ,i out e C denote the flowrate and concentration at the outlet of plant i , respectively. Similarly, out i F and , ,i out e C are the flowrate and concentration at the inlet of plant i , respectively. On the other hand, the concentrations at the outlet and inlet of plant i have an upper limit. max , ,i in e C is the maximum concentration for the inlet of plant i , while max , ,i out e C is the limiting concentration for the outlet of plant i. Mass balance of the separator: ' ' ' ' ' ' , , , , out i itii ir t Ti I r R F F F F i i I t T r R               (8) IiiZFFZF iiiiiiiiii  '' ,,''''' (9) RrIiZFFZF iririririr  ' ,,''''' (10) TtIiZFFZF ititititit  ,, (11) Where out i F denotes the outlet flowrate of water-using plant i . ' ii F is the allocated flowrate from plant i to plant ' i , while ' ir F and it F present the flowrate allocated to the regeneration unit and wastewater treatment unit, respectively. Similar to the mixer, ' ii Z , ' ir Z , it Z is a binary variable denoting the existence of the link between the plant i and water sinks, i.e., plant ' i , regeneration unit and wastewater treatment unit. ''i i F , wr F , ri F are the lower bounds and ''i i F , wi F , ri F are the upper bounds. Due to the limited space, the details of all equations for the regeneration and wastewater treatment plants are not listed here. 2.2 Optimization objective function The objective function of this optimization model seeks to minimize the total cost. The total cost encompasses the costs of freshwater, regeneration, and treatment. The total cost can be written as follows. 142 min COST wi w ri r r r x w W i I r R i I r R F u F u F m                   (12) Where COST is the total cost which is the sum of cost of freshwater, reclaimed water and wastewater treatment. In the right-hand side, wi w w W i I F u       is the total cost of the freshwater consumed by all the plants. Fwi is the flowrate freshwater allocated to plant i and wu is the price of freshwater. ri r r R i I F u       is the total cost for purchasing the regenerated water. Fri is the allocated flowrate from regeneration unit to plant i and r u is the price of regenerated water. The last term in the right-hand side is total cost for wastewater treatment. Fr is the flowrate of wastewater to be treated and rm is the unit cost of wastewater treatment. 3. Case study The case study analysed in this work is reproduced from Keckler and Allen (1999). The relevant limiting data for water intake and effluent discharge of the plants are given in Table 1. This work assumes that there are three plants in the eco-industrial park marked as M, O, and P. The three water contaminants analysed were total organic carbon (TOC), total suspended solids (TSS), and total dissolved solids (TDS). Here, S represents the freshwater suppliers. The relevant costs are shown in Table 2. 3.1 The total water network under limited water supply 3.1.1. Assuming a supply of freshwater F = 1,000 × 10 3 gal/d Scenario a: When the freshwater price U is taken as $ 0.5/10 3 gal, the required minimum consumption of freshwater (Fw) is identified as 1,829 × 10 3 gal/d, i.e., Fw > F. In this scenario, the water supply available is given (F = 1,000 × 10 3 gal/d).Therefore, the actual freshwater consumption is equal to F (1,000 × 10 3 gal/d). Additional regeneration flowrate (829 × 10 3 gal/d) is required. The corresponding total water network is shown in Figure 2. Scenario b: When the freshwater price U is $ 0.75/10 3 gal, the required minimum consumption of freshwater (Fw) is identified as 529 × 10 3 gal/d. In order to meet the water requirements of the three plants, the flowrate of regeneration water should be increased. Compared with Scenario a, the regeneration flowrate of unit B has increased from 829 × 10 3 gal/d to 1,300 × 10 3 gal/d, an increase of 56.8 %. The corresponding total water network is shown in Figure 3. Table 1 Water flowrate and quality requirements for three plants and the supplier water quality Plants Flowrate (10 3 gal/d) Inlets limiting concentration (mg/L) (TOC,TSS,TDS) Outlets limiting concentration(mg/L) (TOC,TSS,TDS) M O P S 11 947 1300 (25,500,2500) (25,25,200) (5,100,500) (1928,2639,7824) (484,105,904) (8,22,276) (0,1,140) Table 2 Freshwater and reused water prices, treatment costs, and outlet concentrations Process Purchase price ($/10 3 gal) Processing costs ($/10 3 gal) Outlets concentration after treatment(mg/L) (TOC,TSS,TDS) A / secondary treatment B / filtration H / buffer tank C / reverse osmosis S / freshwater 0.20 0.25 — 0.30 0.75 5.50 0.40 2.00 6.00 — (20,30,1000) (5,10,500) — (5,1,10) (0,1,140) 3.1.2. Assuming a price of freshwater U fixed at $ 0.8/103 gal Scenario c: When the freshwater supply F is fixed at 2,000 × 10 3 gal/d, the required minimum consumption of freshwater Fw is identified as 529 × 10 3 gal/d. The total water network is similar to that in Figure 2. 143 Scenario d: When the freshwater supply F is fixed at 500 × 10 3 gal/d, the required minimum consumption of freshwater Fw is larger than the supply F. Thus, the minimum consumption of freshwater is chosen as Fw = F = 500 × 10 3 gal/d. In order to meet the needs of three plants, the regeneration units (A/B/C) are all used. The total water network is shown in Figure 4. M O P A H B C D S 529.206 94 7 11 1 1 4 1 7 .7 9 4 87 1. 20 6 9 5 8 1000 To M 470.794 829.206 829.206 M O P A H B C D S 529 94 7 11 1 1 418 87 1 9 5 8 529 1300 1300 To M Figure 2 Total water network at U = $ 0.5/10 3 gal and F = 1,000 × 10 3 gal/d Figure 3 Total water network at U = $ 0.75/10 3 gal and F = 1,000 × 10 3 gal/d 3.2 The relationship between U, F, and Fw Table 3 shows the relevant data describing freshwater price, minimum freshwater flowrate, and regeneration units when freshwater supply F is 2,000 × 10 3 gal/d, 1,000 × 10 3 gal/d, and 500 × 10 3 gal/d, respectively. As shown, for a fixed fresh water supply, its consumption gradually decreases with increasing water price, and more regeneration water units are to be installed. When the constrained water supply is decreased, the corresponding price range of freshwater also decreases and more regeneration water units are used. When the freshwater supply is reduced to 500 × 10 3 gal/d, the regeneration units (A/B/C) are used whichever way prices were presented. When the freshwater price is more than $ 2.57/10 3 gal, the regeneration units (A/B/C) are all used whichever way the freshwater supply were presented. When the freshwater supply is less than the minimum freshwater consumption, the final freshwater supply Fw is equal to F. The relationship between the corresponding price range and the minimum freshwater consumption for various freshwater supplies is presented in Table 4. 4. Conclusions This work could provide technical support for the optimization of inter-plant water integration (IPWI) in industrial parks. In this work, relationships between freshwater price, freshwater supply, minimum freshwater consumption, and the number of wastewater regeneration units were found using mathematical models. When limited boundaries to the freshwater supply were set, the freshwater consumption was found to decrease as freshwater price rises to a certain level. Simultaneously, the number of water regeneration units was found to increase. However, when the freshwater supply was decreased to a particular level, the freshwater price interval was also decreased compared with high water supply scenario, while the number of water regeneration units for this scenario was increased. Table 3 Relationships between F, U, Fw, and regeneration units Freshwater supply F (10 3 gal/d) Freshwater price U ($/10 3 gal) Freshwater flowrate Fw (10 3 gal/d) Regeneration units 2,000 0~0.65 0.65~2.57 >2.57 1,829 529 0 A A/B A/B/C 1,000 0~0.65 0.65~2.57 >2.57 1,000 529 0 A/B A/B A/B/C 500 0~2.57 >2.57 500 0 A/B/C A/B/C 144 Figure 4 Total water network at U = $ 0.8/10 3 gal and F = 500 × 10 3 gal/d Table 4 Relationships between F, U, and Fw Freshwater price U ($/10 3 gal) Freshwater supply F (10 3 gal/d) Freshwater consumption Fw (10 3 gal/d) 0~0.65 F<=1829 F>1829 F 1829 0.65~2.57 F<=529 F>529 F 529 >2.57 F>=0 0 Acknowledgements This work is supported by National Natural Science Foundation of China (nos. 41101570 and 21136003). References Bagajewicz, M., 2000, A review of recent design procedures for water networks in refineries and process plants. Computers and Chemical Engineering, 24(9-10), 2093–2113. Chen C.L., Hung S.W., Lee J.Y., 2010, Design of inter-plant water network with central and decentralized water mains. Computers & Chemical Engineering, 34(9), 1522-1531. Chew I.M.L., Foo D.C.Y., Tan R.R., 2010a, Flowrate Targeting Algorithm for Interplant Resource Conservation Network. Part 1 Unassisted Integration Scheme. Industrial & Engineering Chemistry Research, 49(14), 6439-6455. Chew I.M.L., Foo D.C.Y., Tan R.R., 2010b, Flowrate Targeting Algorithm for Interplant Resource Conservation Network. Part 2: Assisted Integration Scheme. Industrial & Engineering Chemistry Research, 49(14), 6456-6468. Chew I.M.L., Tan R.R., Foo D.C.Y., Chiu A., 2009, Game theory approach to the analysis of inter-plant water integration in an eco-industrial park. Journal of Cleaner Production, 17(18), 1611-1619. Foo, D. C. Y., 2009, State-of-the-Art Review of Pinch Analysis Techniques for Water Network Synthesis. Industrial & Engineering Chemistry Research. 48(11),5125–5159. Jeżowski J., 2010, Review of Water Network Design Methods with Literature Annotations. Industrial & Engineering Chemistry Research, 49(10), 4475–4516. Jia X.P., Liu C.H., Lei S., 2010, Synthesis of Total Water Network based on Water Prices in Eco-industrial Parks. Computers and Applied Chemistry, 27(10), 1449-1452 (In Chinese). Keckler S., Allen D., 1999, Materials reuse modeling A case study of water reuse at an industrial park. Journal of Industrial Ecology, 2(4), 79-92. Lee J.Y., Chen C.L., Lin C.Y., 2013, A Mathematical Model for Water Network Synthesis Involving Mixed Batch and Continuous Units. Industrial & Engineering Chemistry Research, 52(22), 7047-7055. Olesen S.G., Polley G.T., 1996, Dealing with Plant Geography and Piping Constraints in Water Network Design. Process Safety and Environmental Protection, 74(4), 273-276. Pintarič Z.N., Ibrić N., Ahmetović E., Grossmann I.E., Kravanja Z., 2014, Designing optimal water networks for the appropriate economic criteria, Chemical Engineering Transactions, 39, 1021-1026 Rubio-Castro E., Ponce-Ortega J., Serna-González M., El-Halwagi M.M., Pham V., 2013, Global optimization in property-based interplant water integration. AIChE Journal, 59(3), 813-833. Sueviriyapan N., Siemanond K., Quaglia A., GaniR., Suriyapraphadiloket U., 2014, The optimization-based design and synthesis of water network for water management in an industrial process: refinery effluent treatment plant, Chemical Engineering Transactions, 39, 133-138 Tan R.R., Aviso K.B., Cruz J.B. Jr., Culaba A.B., 2011, A note on an extended fuzzy bi-level optimization approach for water exchange in eco-industrial parks with hub topology. Process Safety and Environmental Protection, 89(2), 106-111.