674e Process Debottlenecking: Integrated Techniques for Targeting and Design

Musaed M. Al-Thubaiti, Nasser Al-Azri, and Mahmoud M. El-Halwagi. Artie McFerrin Department of Chemical Engineering, Texas A&M University, College Station, TX 77843

Effective process deottlenecking is one of the most important objectives of the process industries, particularly when there is an additional market demand for the product. This work introduces a novel methodology to debottleneck a chemical process for production maximization. Instead of the conventional unit-by-unit sequential approach for debottlenecking, a new approach is proposed. This new approach is simultaneous in nature and is based on posing the debottlenecking task as a process integration task which links all the design and operating degrees of freedom and exploits synergies among the units and streams to attain maximum debottlenecking. This presentation focuses on no/low cost modifications for process debottlenecking. These include some forms of process intensification as well modest changes in design and operating variables. Process models in the form of path equations are developed to capture the debotllenecking degrees of freedom while describing the process performance with the appropriate level of details. Bounding techniques are used to determine upper bounds on the extent of debottlenecking while local maximization is used to identify lower bounds. A procedure is proposed for the successive iteration and identification of the attainable target of debottlenecking. Next, an optimization formulation is developed and solved to determine the necessary changes needed to reach the target. Specifically, the paper will introduce the following contributions: „X A systematic procedure for the simultaneous debottlenecking is developed. „X A mathematical representation is formulated to characterize the various interactions among the design and operating variables and how they affect the extent of debottlenecking. „X A targeting approach is developed. Because of the non-linearity and non-convexity of most process models, a global bounding technique is employed to determine rigorous targets for debottlenecking. „X An optimization formulation is developed and solved to identify process modifications needed to attain the desired target for debottlenecking. Two case studies is solved to illustrate the applicability of the new approach and its superior results compared to the conventional sequential approach.