Energy Minimization and Design of Heat Exchanger Networks Using Decomposition Strategy

The now well-established pinch design method for the design of energy recovry systems can easily lead to over-integrated heat exchanger (HEN) . Moreover , the inclusion of restrictions as plant layout , match prohibitions and safety considerations into the targeting procedures is not straightforward , laeading to final network designs with a much higher costs than initially estimated . In this work we propose a decomposition strategy where the plant is divided into smaller sub-plants , each containing a subset of the original streams . Each sub-plant is then heat integrated on a stand-alone basis. The streams are randomly distributed among the chosen number of sub-plants , and a simulated annealing optimization procedure is used to identify the distribution of streams . Match restrictions and piping costs due to the excessive distance between two stream can be considered by adding a cost penalty if both streams streams are in the same sub-plant . The objective function is either the total energy requirement or the total annual cost . Once the best configuration is obtained the heat exchanger networks can be designed either manually or using automatic HEN design tools . The latter are now favored due to the amall set off streams in each sub-plant The procedure was applied to a set of plant data , a 9-stream problem and a crude - oil refining process with 24 streams . The results showed that the energy targets for the decomposed problems are similar to the energy requirement of the fully integrated plants , unless the allowed number of streams in each sub-plant is kept very small. Moreover , due the smaller of streams in each set the design of the heat exchanger network is much simpler , and the network costs are very close to the targeted costs . The major benefits are heat recovery systems that are both energy-efficient and simple .