(239e) A Novel Approach for Global Optimization of Stage-Wise Heat Exchanger Network Models

In the last two decades different approaches using mathematical programming were presented for synthesis of heat exchanger networks. Of these, one can classify them as transportation-transshipment oriented and superstructure oriented. Among the superstructure-based models, the most popular method is a stage-wise superstructure approach (Yee and Grossmann, 1990). A comprehensive overview of the advances in this field of research is given in Furman and Sahinidis (2002).

One of the major advantages of the approaches based on mathematical programming is the ability of simultaneous optimization of energy and investment costs. However, this often requires solving a non-convex mixed-integer non-linear problem (MINLP), thus, the solutions obtained may not be globally optimal.

This paper presents a new approach to obtain globally optimal solutions to the Synheat model using recently developed discretization methodology (Bagajewicz and Faria, 2007).

The methodology relies on linearizations of nonlinear equations by discretizing the variables causing nonlinearities over the intervals bounded by their lower and upper bounds. With proper selection of the discrete values, the original non-convex MINLP is converted to a lower-bounding MILP. A proven globally optimal solution is then obtained through the use of an algorithm, which serves to reduce the gap between the lower bounding MILP and original non-convex MINLP.

REFERENCES

Bagajewicz M. J., and Faria D. C., A Discretized Approach to Design Multicomponent Water/ Wastewater Network Systems, AIChE Annual Meeting, Salt Lake City, 2007.

Furman, K. C., Sahinidis, N. V. (2002). A critical review and annotated bibliography for heat exchanger network synthesis in the 20th century. Industrial and Engineering Chemistry Research, 41, 2335–2370.

Yee, T. F., Grossmann, I. E. (1990). Simultaneous optimization models for heat integration II. Computers and Chemical Engineering, 14, 1165–1184.