Global optimization of heat exchanger network synthesis problems by Global Optimum Search (GOS) method

Abstract

The GOS method was applied for solving the heat exchanger network synthesis problems. For solving these problems first, the sources of non-convexities were identified. The variables and constraints were decomposed into two sets. This led to the decomposition of original MINLP problem into two convex subproblems, which can be solved for their respective global solution. Based on the generalized Bender's decomposition method, this approach then solved a series of these subproblems to arrive at the optimal solution. Even though the proposed approach cannot mathematically guarantee the identification of the global optimum, the global optimization technique applied to a number of HENS problems determined the global solution in all cases from various starting points. The mathematical modeling of heat exchanger network synthesis and the modified global optimum search approach was implemented automatically using the modeling language GAMS. This is an abstract of a paper presented at the 18th International Congress of Chemical Process Engineering (Praque, Czech Republic 8/24-28/2008).