A Systematic State-Space Superstructure Based Methodology for Batch Mass Exchange Network Design
Abstract
In the literature, studies on batch mass exchange network (MEN) design are relatively scarce and late in contrast to MEN of continuous processes, though batch productions own incomparable advantages in producing small-amount and high-value-added products. In batch MEN synthesizing, matches between rich and lean process streams are characterized not only by flow rate and concentration, but also time, so that storage strategies are always adopted for material-reusing purpose among different time periods. In a typical batch MEN, the valuable or undesirable components in rich process streams can be reduced in countercurrent direct-contacted mass exchanges by contacting with lean process streams or external mass separating agents (MSAs) in each time interval. The main objectives of batch MEN design are minimum consumption of (external) MSAs/discharge of wastes, or the total cost.
In the past, both the insight-based graphical pinch analysis and the mathematical-based optimization techniques have been adopted in batch MEN studies, but none of these works (C. Y. Foo et al. 2004, 2005; C. L. Chen and Y. J. Ciou 2006, 2007) can obtain the minimal total cost in one step. Furthermore, these works also failed to provide opportunities for non-isocomposition mixing/splitting and the locations for storage tanks are limited in the component dimension.
This paper aims at developing a systematic methodology to optimize operating cost and capital cost of batch mass exchange networks in a simultaneous manner. Specifically, the traditional state space superstructure is divided into several "full mesh" subsystems through the property-based mixing/splitting rules (Figure 1), to capture the potential connecting and matching structures while eliminating all the unnecessary connections. The proposed superstructure can easily handle non-isocomposition mixing/splitting as well as flexible storage locations both in the component and time gamut. Then, a mixed-integer nonlinear program (MINLP) is first formulated to minimize the total annual cost (TAC), in which the balance between operating cost (utility cost) and capital cost (investment on operating units and storage tanks) can be achieved in one step. Furthermore, the multi-purpose units and parallel operations are also introduced not only to enhance the flexibility of batch process production, but also to make a better use of all the equipment and stream resources to further explore the benefits in both space and time dimensions. Finally, the coke oven gases (COG) separating problem is presented to demonstrate the applicability and superiority the proposed approach for batch MEN design. It turned out the TAC can be reduced by 5-70 percent compared with that of the previous works, and another over 10 percent can be decreased by using the multi-purpose units and parallel operations.
Figure 1 The improved State-Space
Superstructure for batch MEN design Keywords: Batch
mass exchange networks; State Space superstructure; MINLP; Multi-purpose units;
Parallel operations References [1]
Foo C. Y.; Manan Z. A.; Yunus R. M.; Aziz R. A. Synthesis of mass exchange
network for batch processessPart I: Utility targeting. Chem. Eng. Sci., 2004,
59, 1009¨C1026. [2]
Foo C. Y.; Manan Z. A.; Yunus R. M.; Aziz R. A. Synthesis of mass exchange
network for batch processes Part II: Minimum units target and batch network
design. Chem. Eng. Sci. 2005, 60, 1349¨C1362. [3]
Chen, C. L.; Ciou, Y. J. Superstructure-Based MINLP Formulation for Synthesis
of Semicontinuous Mass Exchanger Networks. Ind. Eng. Chem. Res. 2006, 45, 6728¨C6739. [4]
Chen, C. L.; Ciou, Y. J. Synthesis of a Continuously Operated Mass-Exchanger
Network for a Semicontinuous Process. Ind. Eng. Chem. Res. 2007, 46, 7136¨C7151.
See more of this Group/Topical: Computing and Systems Technology Division