(583g) Optimal Mass Exchanger Network Synthesis Using a 2-Step Hybrid Algorithm Including Packed Column Design | AIChE

(583g) Optimal Mass Exchanger Network Synthesis Using a 2-Step Hybrid Algorithm Including Packed Column Design

Authors 

Short, M. - Presenter, Carnegie Mellon University
Biegler, L., Carnegie Mellon University
Isafiade, A. J., University of Cape Town

A method for the synthesis of mass exchanger networks (MENs) is presented that focuses on including detailed packed column design considerations at the network optimization level to deliver designs that are optimal and more realistic. Most approaches to simultaneous optimization of MENs have been based upon a stage-wise superstructure with simplified representations of the individual mass exchangers. Even with these simplified representations, the MINLP is still non-convex and computationally difficult to solve with modern deterministic solvers. Additionally, it is not possible to guarantee that when a solution is found that it is globally optimal.

The method developed in this study, uses a version of this aforementioned stage-wise MINLP formulation in the first step. Once the solution to this is found, a secondary nonlinear programming (NLP) step, based on orthogonal collocation on finite elements, is performed to find the optimal column dimensions, packing sizes, fluid velocities, and pressure drops, based on the flows and stream matches from the network optimization. Once the new individual column designs are found, the MINLP is then updated with a series of correction factors based on the solution of the detailed designs. This procedure repeats between the MINLP and NLP sub-optimization until the differences between the solutions converge or until a maximum number of iterations is reached.

The methodology, presented in detail in the work of Short, et al. (2018), is moved from the GAMS environment into PYOMO, allowing for powerful new ways to enhance the algorithm. The previous methodology’s correction factor determination is improved upon in order to speed up and increase the chances of convergence. Additionally, through the application of different MINLP solvers, larger examples that were previously not able to be solved are included.

Reference: Short, M., Isafiade, A.J., Biegler, L.T.,Kravanja, Z., 2018, Synthesis of mass exchanger networks in a two-step hybrid optimization strategy, Chemical Engineering Science, 178, 118 - 115.

Checkout

This paper has an Extended Abstract file available; you must purchase the conference proceedings to access it.

Checkout

Do you already own this?

Pricing

Individuals

AIChE Pro Members $150.00
AIChE Graduate Student Members Free
AIChE Undergraduate Student Members Free
AIChE Explorer Members $225.00
Non-Members $225.00