(246a) A Simultaneous Utility and Area Targeting Model for Integrated Process and Heat Exchanger Network Synthesis

Over the past four decades, the field of heat integration has flourished and a large body of literature regarding heat exchanger network synthesis (HENS) has been published. Traditional HENS methods assume that the reaction and separation networks are designed first, resulting in fixed stream temperatures and flowrates which are then used for the synthesis of the heat exchanger network (HEN) in a three-step procedure: (1) minimize utility consumption, (2) minimize number of exchanger units, and (3) minimize area and obtain HEN. It is known that sequential methods do not fully consider the interactions among steps, which can lead to suboptimal solutions. To address this, some works allow simultaneous HENS that accounts for the trade-offs between operating cost and capital investment [1, 2], while others incorporate utility targeting with process synthesis to account for interaction between process and HEN [3, 4]. However, no previous work has simultaneously considered capital and operating cost in both chemical process and heat exchanger network.

Accordingly, in this work, we present a mixed-integer nonlinear programming (MINLP) model for simultaneous utility and area targeting for integrated process and HEN synthesis. The model is built upon: (1) the NLP model by Colberg and Morari [5] that allows simultaneous HENS with fixed stream temperature and flowrate, and (2) the simultaneous process synthesis and utility targeting MINLP model by Kong et al. [6]. Incorporating and extending the ideas from these two approaches, here we propose a transshipment-based model in which the temperature intervals are implicitly constructed via a dynamic temperature grid so that variable stream temperature and flowrate can be handled. Using binary variables, the stream inlet and outlet temperatures are properly mapped to the grid, and thus the stream heat duties in each interval can be determined. We introduce heat cascades for both the hot and cold streams, and calculate “deflected” heat for cold streams and residual heat for hot streams. Using the residual and deflected heats, the inlet/outlet temperatures of each individual exchanger in each interval are determined. Finally, area and cost of each exchanger can be calculated through log-mean temperature differences and exchanger heat loads.

In each interval, should stream splitting is present the flowrate and exchanger heat load at each branch are chosen so that isothermal mixing is achieved. The resulting HEN in each interval has heat exchangers placed in parallel. In general, the number of grid points required is equal to the total number of streams (including utilities). Introducing additional grid points might lead to improved designs, albeit at the cost of computational performance. We present constraints tailored to help finding the globally optimal HEN configuration. For example, if the same pair of hot and cold stream exchange heat in two consecutive intervals, there will only be one heat exchanger that includes the combined heat exchanged in these two intervals.

We show how the model can handle constrained (i.e. required or forbidden) matching, isothermal and non-isothermal streams, and multiple utilities. The model can also be extended to handle streams that cannot be classified as hot or cold a priori. In addition, we introduce a preprocessing algorithm to generate tight variable bounds and to eliminate sub-optimal solutions.

The model is first compared with state-of-the-art simultaneous HENS strategies through examples with fixed temperatures and flowrates. Then, it is applied on several illustrative examples with variable temperatures and flowrates. Finally, we present a case study in which the proposed model is integrated with a realistic superstructure-based process synthesis model.

References

1. Ciric, A.R. and C.A. Floudas, Heat exchanger network synthesis without decomposition. Computers & Chemical Engineering, 1991. 15(6): p. 385-396 DOI: Doi: 10.1016/0098-1354(91)87017-4.

2. Yee, T.F. and I.E. Grossmann, Simultaneous optimization models for heat integration--II. Heat exchanger network synthesis. Computers & Chemical Engineering, 1990. 14(10): p. 1165-1184 DOI: Doi: 10.1016/0098-1354(90)85010-8.

3. Navarro-Amorós, M.A., et al., An alternative disjunctive optimization model for heat integration with variable temperatures. Computers & Chemical Engineering, 2013. 56: p. 12-26

4. Duran, M.A. and I.E. Grossmann, Simultaneous optimization and heat integration of chemical processes. AIChE Journal, 1986. 32(1): p. 123-138 DOI: 10.1002/aic.690320114.

5. Colberg, R. and M. Morari, Area and capital cost targets for heat exchanger network synthesis with constrained matches and unequal heat transfer coefficients. Computers & chemical engineering, 1990. 14(1): p. 1-22

6. Kong, L., et al., Simultaneous chemical process synthesis and heat integration with unclassified hot/cold process streams. Computers & Chemical Engineering, 2017. 101: p. 210-225