Optimization of a saturated gas plant: Meticulous simulation-based optimization - A case study.

Area of Science:

• Chemical Engineering
• Process Optimization
• Computational Simulation

Background:

• Literature often optimizes individual units or discrete design choices in process simulation.
• Scalable continuous design variable optimization is crucial for industrial applications.
• Complex, multi-component systems require robust optimization strategies.

Purpose of the Study:

• To develop and apply a simulation-based optimization strategy for a full-scale green-field saturated gas plant.
• To compare the efficacy of genetic algorithm (GA) and particle swarm optimization (PSO) for complex process optimization.
• To minimize total annual cost while guaranteeing a minimum butane recovery rate.

Main Methods:

• Coupling Aspen HYSYS® (process simulator) with MATLAB® (programming tool).
• Implementing a hybrid strategy combining sensitivity analysis, constrained bounding, and stochastic optimization (GA, PSO).
• Performing steady-state simulation-based optimization on a real-world, multi-component gas plant.

Main Results:

• Both GA and PSO successfully optimized the gas plant's operating conditions.
• GA demonstrated greater confidence and handled wider constrained bounds effectively.
• The strategy minimized total annual cost, achieving at least 92% butane recovery.

Conclusions:

• The developed optimization-simulation strategy is effective for complex industrial processes.
• This approach is recommended for optimizing operating conditions in existing (brownfield) plants due to feedstock variability.
• Simulation-based optimization offers a powerful tool for enhancing efficiency and profitability in the gas processing industry.