Conceptual methods for synthesis of reactive distillation processes: recent developments and perspectives

Reactive distillation (RD) is a process intensification technique that offers major advantages over conventional technologies by enabling the integration of reaction and separation into a single apparatus. This integration introduces and exploits complex interaction between mass transfer, chemical reaction and hydrodynamics within an RD...

Operating windows for early evaluation of the applicability of advanced reactive distillation technologies

Advanced reactive distillation technologies (ARDT) are often overlooked during process synthesis due to their complexity. This work proposes the use of operating windows with additional features to identify suitable operating limits for ARDT. Data needed to construct the operating windows are thermodynamic properties, kinetic parameters,...

Synthesis and optimization of energy integrated advanced distillation sequences

This paper explores the basis on which reliable screening of distillation sequences for energy-efficient separation of zeotropic multicomponent mixtures can be carried out. A case study for the separation of natural gas liquids is used to demonstrate the approach. To solve this generic problem, a screening algorithm has been developed using...

Synthesis and optimization of NGL separation as a complex energy-integrated distillation sequence

The synthesis of heat-integrated distillation sequences for energy-efficient separation of zeotropic multicomponent mixtures is complex due to the many interconnected design degrees of freedom. This paper explores the basis on which reliable screening can be carried out. To solve this problem, a screening algorithm has been developed using...

Flowsheet generation through hierarchical reinforcement learning and graph neural networks

Process synthesis experiences a disruptive transformation accelerated by artificial intelligence. We propose a reinforcement learning algorithm for chemical process design based on a state-of-the-art actor-critic logic. Our proposed algorithm represents chemical processes as graphs and uses graph convolutional neural networks to learn from...