(185h) Tackling the Challenges/Limitations Posed By Heat Exchanger Network in Work-Heat Exchange Network Synthesis

The recent concerns of global warming and exhausting natural resources have encouraged global efforts for increased energy efficiency and conservation. As industry is one of the major energy consumer, tools and methods are being developed and applied to reduce the overall industrial energy consumption. In process industries, heat and work are the two main forms of energy consumption. Heat integration is a well-established tool in process industries to reduce the overall utility consumption. These principles of heat integration can also be extended to work integration. Furthermore, considerable energy savings can be achieved by simultaneously integrating both forms of energy (work and heat) requirements of the process streams.

There are novel techniques available to incorporate the heat integration model in process synthesis and optimization. Duran and Grossmann [1] developed the Pinch Location method for obtaining the minimum utilities within process flowsheet optimization. Kong, et al. [2] developed a generalized transshipment model for unclassified hot/cold streams for simultaneous process synthesis and heat integration. However, the focus of these heat integration models is to obtain the minimum utility consumption rather than designing the network.

In work-heat exchange network synthesis (WHENS), a superstructure for designing heat exchanger network (HEN) is incorporated. This embedded HEN superstructure is of importance. The overall mathematical complexity of the synthesis might increase with the complexity of this HEN superstructure. As a result, simpler HEN superstructures with simplified assumptions are used to reduce the complexity and computation time of the model. These superstructures have some limitations associated with it, which reflects in the final result of WHENS. This may eliminate many industry relevant promising configurations and operating conditions. For example, some of the commonly used superstructures allow utility exchangers only at the end of superstructure. Furthermore, some of the assumptions in the HEN model such as isothermal mixing might eliminate simpler and optimal configurations. One more important challenge is that the identity of the process streams as hot/cold is not known beforehand in WHENS. Hence, it is difficult to classify the streams and enumerate the matches in the HEN superstructure. Thus, novel methods and strategies are required to tackle these challenges and obtain a reasonable solution for the entire problem.

In this work, we consider superstructure based simultaneous WHENS. We have developed a novel framework for work-heat integration in which we do not fix the identity of the streams as hot or cold a priori in heat integration [3]. Furthermore, we used zone-based pressure-temperature dependent properties, allowing liquid, vapor, and 2-phase streams. However, this mixed-integer non-linear programming (MINLP) model uses stage-wise superstructure [4] with isothermal mixing assumption for heat integration. For a case study of C3 splitter, we observe that the reboiler has a utility exchanger in series rather than the preferred and more practical parallel configuration. Thus, we evaluate the effect of the HEN superstructure and model assumptions on such embedded process synthesis and implement novel strategies for a case study of C3 splitter. Overall, we improve the HEN superstructure and relax certain assumptions to improve the result of WHENS.

[1] M. A. Duran and I. E. Grossmann, "Simultaneous optimization and heat integration of chemical processes," AIChE Journal, vol. 32, pp. 123-138, 1986.

[2] L. Kong, V. Avadiappan, K. Huang, and C. T. Maravelias, "Simultaneous chemical process synthesis and heat integration with unclassified hot/cold process streams," Computers & Chemical Engineering, vol. 101, pp. 210-225, 2017.

[3] S. K. Nair, H. N. Rao, and I. A. Karimi, "Framework for work‐heat exchange network synthesis (WHENS)," AIChE Journal.

[4] T. F. Yee and I. E. Grossmann, "Simultaneous optimization models for heat integration--II. Heat exchanger network synthesis," Computers and Chemical Engineering, vol. 14, pp. 1165-1184, 1990.