Optimal Operation of an Industrial Dividing Wall Column through Multiparametric Programming

Area of Science:

• Chemical Engineering
• Process Control
• Separation Technology

Background:

• Dividing wall columns (DWCs) offer energy and capital savings in chemical separations.
• Accurate dynamic modeling is crucial for effective control of complex distillation processes.
• Methyl methacrylate separation presents challenges due to multi-component mixtures and non-ideal thermodynamics.

Purpose of the Study:

• To develop a high-fidelity dynamic model of an industrial dividing wall column for methyl methacrylate separation.
• To design and implement an explicit model predictive controller ( EMPC) for process regulation.
• To demonstrate the effectiveness of the proposed modeling and control strategy for optimal online operation.

Main Methods:

• Development of an equation-oriented mathematical model incorporating non-ideal thermodynamic calculations.
• Synthesis and application of an explicit model predictive controller ( EMPC).
• Simulation and analysis of the dividing wall column coupled with a decanter.

Main Results:

• A detailed, high-fidelity dynamic model of the dividing wall column process was successfully developed.
• The explicit model predictive controller effectively tracked product purity specifications.
• The coupled system achieved highly concentrated methyl methacrylate and water streams.

Conclusions:

• The developed dynamic model accurately represents the industrial dividing wall column.
• Explicit model predictive control enhances process operability and ensures product quality.
• The combined modeling and control approach enables optimal online operation for methyl methacrylate separation.