Bi-level optimizing control of a simulated moving bed process with nonlinear adsorption isotherms

Area of Science:

• Chemical Engineering
• Separation Science
• Process Control

Background:

• Simulated Moving Bed (SMB) chromatography is crucial for industrial separations.
• Existing control schemes often assume linear isotherms, limiting performance with complex mixtures.
• Nonlinear isotherms present significant challenges for SMB process control and purity regulation.

Purpose of the Study:

• To extend a bi-level optimizing control scheme to SMB processes with nonlinear isotherms.
• To evaluate the performance of the enhanced control scheme against disturbances.
• To compare the proposed scheme with a control strategy based on linear isotherms.

Main Methods:

• A bi-level optimization framework was implemented, featuring cyclic steady-state optimization in the upper level and repetitive model predictive control in the lower level.
• The control scheme utilized time-varying flow rates and switching periods, with decision variables from the upper level informing the lower level.
• Experimental and numerical studies were conducted on a high-concentration mixture of L-ribose and L-arabinose.

Main Results:

• The proposed bi-level control scheme demonstrated satisfactory performance and robust purity regulation for SMB separation of nonlinear mixtures.
• The advanced scheme effectively handled various process disturbances.
• An alternative control scheme based on linear isotherms exhibited limitations in purity regulation, failing under similar conditions.

Conclusions:

• The extended bi-level optimizing control scheme is effective for simulated moving bed chromatography with nonlinear isotherms.
• This approach offers superior control performance and reliability compared to methods relying on linear isotherm assumptions.
• The study highlights the importance of accounting for nonlinearities in SMB process control for challenging separations.