Optimization and network sensitivity analysis for process retrofitting

Retrofitting is the redesign of an operating chemical plant to find new configurations and optimal operating conditions. In the chemical industry, 60% of new capital investments in plants and equipment are retrofitting projects, while only 10% goes to building new plants. Investment in retrofitting amounted to $26 billion in 2000. Despite the importance of retrofitting, there are few methodologies for finding improved economic and environmental performance for continuous processes. This work proposes a systematic framework for the understanding of retrofitting of continuous chemical processes and develops a new methodology to support decision making in solving this problem. Successful retrofitting solutions derive from a balance of operational experience in the plant and the rigor of mathematical analysis. This balance is accomplished by proposing tools and algorithms that in the problem formulation, the analysis of the flowsheet, the synthesis of retrofitting options and the final decision, allow the decision maker to handle the complexity of the problem and focus on the truly critical aspects of the flowsheet. The proposed methodology structures the problem by defining a broad range of retrofitting objectives and alternatives. The initial step is the formulation of retrofitting as an optimization problem. This includes defining retrofitting goals and translating them into objective functions. A parameter optimization of the base case design determines the incentives and constraints for retrofitting. The analysis continues through a network optimization analogy. The representation of the flowsheet as a multicommodity network allows the use of a graph based algorithm to determine the cycles in the process and apply flow decomposition by techniques developed in this study. Flow decomposition determines the path and cycles by which commodities (chemicals) flow through the network. The focus on chemicals and their paths rather than unit operations avoids the distinction of process subsystems providing an integrated view of the flowsheet. The objective function is evaluated in terms of path and cycle flows. Using graphical and mathematical programming (sensitivity analysis) approaches, the synthesis stage identifies retrofitting opportunities that increase the favorable and limit the unfavorable paths and cycles. Once a set of appropriate retrofitting alternatives is identified. the decision stage proceeds through a systematic construction of the superstructure and the corresponding MINLP model. The procedure takes into account the implicit logic of the retrofit alternatives to reduce the space of decision variables. The methodology is completed with a framework to implement the outer approximation algorithm taking into account the characteristics of the retrofitting problem. Case studies illustrate the benefits of the different stages of the proposed retrofitting methodology: efficient solution algorithms, systematic ways to analyze and generate alternative plant configurations and ease in finding optimal designs and investment decisions. The new methodology is compatible with existing flowsheet simulation tools and optimization packages and can easily be applied to a wide range of practical problems.

Department dc:contributor.department

Massachusetts Institute of Technology. Dept. of Chemical Engineering.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2001

Author dc:creator

• Guinand, Ernique A. (Ernique Alberto), 1970-

Advisor dc:contributor.advisor

• Gregory J.McRae.

• Chemical Engineering.

Statement dc:rights

• M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/8744

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/8744