| dc.description.abstract | Process intensification (PI) addresses the development of new equipment and processing techniques resulting in substantially smaller, cleaner, safer, and more energy-efficient technologies. It is a novel design concept suggesting an innovative outlook towards chemical processes. Incorporation of PI principles into the conceptual process design stage, where the initial layout of the plant is decided, can be beneficial not only in terms of economics, but it can also help to mitigate the industrial footprint on global warming and environmental pollution. However, identification of such intensified solutions at the conceptual design stage is a challenging task as there can be myriad of candidate process configurations. While optimization-based process synthesis approaches provide methodical tools for process design, they request pre-postulated superstructures with fixed connectivity and equipment types. This limits the scope for the discovery of unconventional design solutions. In this work, we present a new representation method for chemical processes based on building blocks which enables an optimization-based approach for systematic intensification of chemical processes. Building block-based representation does not require a priori postulation of equipment types and configurations and allows for a systematic representation, identification and generation of intensification alternatives at the equipment and flowsheet levels. The proposed approach not only identifies different intensified/traditional process equipment, but also automatically generates the corresponding flowsheet. Overall problem is formulated as a Mixed-Integer Nonlinear Optimization Problem (MINLP) where discrete variables are used in selection of the phenomena and enabling materials. Proposed superstructure representation is also generalized into a unified framework for solving different process synthesis and integration problems with a single superstructure eliminating the need for postulating new superstructures whenever a new problem is addressed. Several solution strategies are devised to address the solution of the MINLP model by exploiting the special structure of the representation. The capabilities of the proposed method are demonstrated through a wide range of examples. | en |
