Available online at www.sciencedirect.com Computers and Chemical Engineering 32 (2008) 540–551 Simultaneous targeting and design for cooling water systems with multiple cooling water supplies Thokozani Majozi a,b,∗ , Anand Moodley b a Department of Computer Science, University of Pannonia, Egyetem u. 10, Veszpr´ em H-8200, Hungary b Department of Chemical Engineering, University of Pretoria, Lynnwood Road, Pretoria 0002, South Africa Received 23 November 2006; received in revised form 2 March 2007; accepted 21 March 2007 Available online 27 March 2007 Abstract This paper presents a technique for simultaneous targeting and design in cooling water systems comprising of at least two cooling towers and several cooling water using operations. The presented technique is based on a superstructure from which a mathematical formulation is derived using system specific variables and parameters. It is demonstrated that in a system like this, true optimality can only be realized by a holistic consideration of the entire cooling water system. Consideration of individual subsets of cooling towers with their dedicated cooling water operations yields suboptimal results. Four operational cases are considered and structural considerations of corresponding mathematical formulations presented. The first case results in a linear programming (LP) formulation, the second case yields a mixed integer linear programming (MILP) formulation whilst the other two cases yield mixed integer nonlinear programming (MINLP) formulations which cannot be exactly linearized. However, in all cases significant improvements in excess of 40% were realized in targeting, without compromising the heat duty of the cooling water using operations. The main objective of this investigation is to debottleneck the overall cooling water supply for the cooling water network. © 2007 Elsevier Ltd. All rights reserved. Keywords: Design; Targeting; Cooling tower; Mathematical optimization; Optimality 1. Introduction There are inherently large quantities of energy and water used in chemical processes. Since the late seventies, significant efforts have been put in developing systematic procedures for energy optimization in grassroot and retrofit design. This resulted in the development of pinch analysis (Linnhoff, Mason, & Wardle, 1979), which has been successfully applied to various indus- tries worldwide. This successful development was followed by the development of similar techniques for water minimiza- tion in chemical processing industries from the late eighties to very recently (Alva-Arg´ aez, Kokossis, & Smith, 1998; Doyle & Smith, 1997; El-Halwagi, El-Halwagi, & Manousiouthakis, 1992; Hallale, 2002; J¨ odicke, Fischer, & Hungerb¨ uhler, 2001; Olesen & Polley, 1997; Savelski & Bagajewicz, 2000; Wang & Smith, 1994, 1995a, 1995b). Whilst this technology has not fully ∗ Corresponding author at: Department of Chemical Engineering, University of Pretoria, Lynnwood Road, Pretoria 0002, South Africa. Tel.: +27 12 420 4130; fax: +27 12 362 5173. E-mail address: thoko.majozi@up.ac.za (T. Majozi). developed, various industries, globally, have started reaping the benefits of its application. The main drawback of the previous developments is that mass and heat integration problems are treated in a dichotomous manner. Also, significant work has been done on cooling water sys- tems relating to the reliability of cooling towers (Bartoli, Borri, & Zahlten, 1992; Kasperski & Niemann, 1988; Sudret, Defaux, & Pendola, 2005; Wittek & Meiswinkel, 1998), the optimum sizing of cooling towers (Soylemez, 2001), energy conserva- tion (Knoche and Bosnjakovic, 1998), cooling water treatment (Martinez, Gallegos, & Martinez, 2004) and other operational issues of cooling water systems. These developments do not con- sider the impact of the cooling water network on the performance of the cooling tower. It is only recently that combined heat and mass integra- tion have been studied. Kim and Smith (2001) have developed and applied a graphical technique for maximization of cool- ing tower performance through minimization of supply water to the cooling water network. The cooling water network involves all cooling water using operations including heat exchangers. This methodology demonstrated that cooling water minimiza- 0098-1354/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.compchemeng.2007.03.016