Computers and Chemical Engineering 37 (2012) 104–118 Contents lists available at SciVerse ScienceDirect Computers and Chemical Engineering j ourna l ho me pag e: w ww.elsevier.com/locate/compchemeng Energy integration of industrial sites with heat exchange restrictions Helen Becker ∗ , Franc ¸ ois Maréchal Industrial Energy Systems Laboratory, Ecole Polytechnique Fédérale de Lausanne, CH-1015, Switzerland a r t i c l e i n f o Article history: Received 14 June 2011 Received in revised form 25 August 2011 Accepted 20 September 2011 Available online 6 October 2011 Keywords: Pinch analysis Utility integration Restricted matches Process sub-system Envelope composite curves Heat load distribution a b s t r a c t Process integration methods aim at identifying options for heat recovery and optimal energy conver- sion in industrial processes. This paper introduces a targeting method, which includes heat exchange restrictions between process sub-systems. The problem is formulated as a MILP (mixed integer linear programming) problem, which considers not only restricted matches but also the optimal integration of intermediate heat transfer units and the energy conversion system, like heat pumping and combined heat and power production. Moreover a new mathematical formulation is presented to chose optimal heat transfer technologies. For solutions avoiding the energy penalty, the composite curves of optimal heat transfer units have to be embedded between the new generated hot and cold envelope composite curves. The application of the method is illustrated through an industrial example from the pulp and paper industry. © 2011 Elsevier Ltd. All rights reserved. 1. Introduction Pinch analysis is a promising tool to optimize the energy efficiency of industrial processes. To realize the maximum heat recovery and the optimal integration of utilities to supply process heating and cooling requirements, first the heat load distribution based on process and optimal utility streams has to be calculated. One major difficulty is the assumption that any hot stream can exchange heat with any cold stream. In reality, heat exchanges become difficult or even impossible, due to constraints such as the distance between streams or product quality and/or safety reasons, or due to system dynamics such as non-simultaneous operations. Forbidden matches between certain pairs of process streams are considered by Papoulias and Grossmann (1983). They propose a mathematical formulation to identify the heat load distribution that minimizes the energy penalty of restricted matches without proposing any solutions for adding heat transfer fluids or integrat- ing utility systems. Also Cerda and Westerberg (1983) studied heat exchanger networks with restricted matches and propose an algo- rithm which imposes constraints disallowing in part or in total the matching of stream pairs. The total site approach, presented by Dhole and Linnhoff (1992) and later by Klemeˇ s, Dhole, Raissi, Perry, and Puigjaner (1997), implicitly accounts for restricted matches before designing the heat exchanger network. The hot and cold streams, resulting from sub-systems without considering self-sufficient pockets, are ∗ Corresponding author. Tel.: +41 21 693 3550; fax: +41 21 693 3502. E-mail address: helen.becker@epfl.ch (H. Becker). separated graphically. The sub-systems can only exchange heat via the steam system. Also Hui and Ahmad (1994) studied total site integration with indirect heat transfer between process plants through steam utilization from the steam network. In this work exergy analysis is used and the self-sufficient zones were not always suppressed. More directly, Rodera and Bagajewicz (1999) pointed out that skipping the self sufficient pocket can reduce significantly the opportunities for heat recovery and they present a transship model which calculates the heat to be transferred between two process plants. An extension to several plants is proposed later by the same authors (Bagajewicz & Rodera, 2000, 2002). Bagajewicz and Rodera (2001) propose a single heat belt, which exchanges heat between process plants by an intermediate fluid. Only for special cases (3 process plants) this problem can be solved with a MILP formulation. Combining the total site proposed by Dhole and Linnhoff (1992) and the approach of Bagajewicz and Rodera (2000), Bandyopadhyay, Varghese, and Bansal (2010) introduces site level grand composite curves for indirect heat transfer. Indirect heat transfer between plants and an extension to industrial zones containing several process plants is presented by Stijepovic and Linke (2011). Mainly the utility system is optimized and only waste heat can be transferred between process plants. Maréchal and Kalitventzeff (1999) propose a MILP strategy, which integrates forbidden heat exchange connections as con- straints in the targeting phase, and allows the integration of heat transfer fluids. The penalty in terms of utility and operating costs can be considered. This paper proposes an extension of the MILP strategy and, depending on a given process or system, a systematic approach to define the members of sub-systems. Heat exchange between 0098-1354/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.compchemeng.2011.09.014