(112d) A 'valve-Free' Model for Dynamic Simulation of Pressure Swing Adsorption Processes That Automatically Determines Near-Optimal Operational Policies

Kim, T., Georgia Institute of Technology
Scott, J., Georgia Institute of Technology
Sees, M., Texas Tech University
Chen, C. C., Texas Tech University
Kirkes, T., Texas Tech University
In this oral presentation, we will present a new 'valve-free' modeling framework for pressure swing adsorption (PSA) processes that automatically determines near-optimal operational decisions during dynamic simulation. More specifically, the proposed method eliminates the need for the user to specify inputs concerning the detailed dynamic operation of the PSA cycle, thereby enabling non-expert users to obtain simulation results corresponding to near-optimal operation of the PSA cycle. This methodology is particularly useful for adsorbent screening, where making meaningful comparisons between adsorbents is difficult using standard dynamic simulation techniques because the optimal operational decisions are highly dependent on adsorbent properties.

The development of a wide variety of novel adsorbents, such as the general classes of zeolites, metal organic frameworks (MOFs), and zeolitic imidazolate frameworks (ZIFs), raises new opportunities for improving the performance of PSA processes in terms of product purity, recovery, productivity, energy efficiency, and capital cost. However, designing an economical and efficient PSA process for a novel adsorbent may require an identification of the right adsorbent for a given application. This highlights the need for effective in silico material screening tools that can screen sub-optimal adsorbent candidates asides so that any improvement activities can be focused on adsorbents with the biggest potential impacts on PSA process performance.

Currently, most adsorbent material screening methods focus on adsorbent material properties such as adsorption capacity and product selectivity. However, restricting attention solely to the material properties effectively ignores critical features of the overall adsorption dynamics in PSA processes. In terms of adding another layer of screening for these adsorbent materials based on the process system engineering (PSE) considerations, dynamic simulations for PSA processes can exclusively be utilized due to their rigorous simulation capabilities through high-fidelity dynamic model. Furthermore, such dynamic process simulations enable the user to be able to perform high throughput screening of novel adsorbents with minimal concerns in constraints in time and capital investments.

While there have been several studies on adsorbents screening based on dynamic simulation of PSA processes, conducting such studies still requires the solution of complex design, control, and optimization problems given each adsorbent; these approaches may require significant expertise from the practitioner, thus difficult to be widely accepted by the practitioners. Not only that, the state-of-the-art dynamic process simulation based design of PSA processes often requires trial and error approaches thus can easily be lead to unreliable results. In this oral presentation, we would like to formulate and introduce a solution to the identified limitations of the current state-of-the-art by providing a cutting-edge method for providing near-optimal operational policies for PSA processes.

In order to address the aforementioned limitations, we will introduce a new and simplified valve-free PSA process dynamic model that enables more facile design of PSA processes. The model utilizes input parameters that are more intuitive in their meanings so that the user can specify them with more confidence; also the model inputs are reduced in their numbers thus lowering a probability of running unreliable simulations. Furthermore, the model allows the user to easily configure near-optimal PSA processes through high fidelity dynamic simulations by automating key operational policies.

Through the presentation, not only we will demonstrate applicability of our model in dynamic simulation of PSA processes but also show that our model would accept much more universally applicable choices of input parameters in assessing vastly different adsorbent candidates. Finally, we would like to highlight the fact that the dynamic simulation with valve-free model can bring insights to the practitioners regarding near-optimal operations of PSA processes.


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


Do you already own this?



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