This paper proposes an integrated framework that consistently links design, operation, and control for membrane-based gas separation. First, a two-dimensional transport model quantifies how wetting modifies membrane transfer resistance, showing that t...
This paper proposes an integrated framework that consistently links design, operation, and control for membrane-based gas separation. First, a two-dimensional transport model quantifies how wetting modifies membrane transfer resistance, showing that the liquid-phase diffusion layer dominates the overall mass-transfer resistance and that partial wetting markedly suppresses effective flux. Although polypropylene (PP) exhibits slightly superior dry performance, polymethylpentene (PMP) is less susceptible to wetting and pore intrusion, thereby offering higher long- term stability. Building on these transport insights, we construct a steady-state, one- dimensional absorption–stripping process model for isopropanol (IPA) removal using a hollow fiber membrane (HFM) module and perform cost minimization under recovery and purity constraints that captures the tradeoff between membrane area and utility consumption. As the liquid flow rate increases, enhanced mass transfer reduces capital expenditure (CAPEX), whereas higher vacuum and circulation-pump loads increase operating expenditure (OPEX), yielding a characteristic decrease- then-increase in total cost with a turning point near 2,171 kg/min. Finally, based on the optimal design, we develop a one-dimensional dynamic model and an estimation- and-control scheme combining moving-horizon estimation (MHE) with model predictive control (MPC). The proposed MHE–MPC maintains product CH₄ purity within constraints under feed-flow and composition disturbances and measurement noise, and it outperforms a conventional proportional–integral–derivative (PID) controller in setpoint tracking, robustness, and disturbance rejection. By connecting (i) a wetting-aware mechanistic model, (ii) process techno-economic optimization, and (iii) closed-loop operation, the framework provides a species-agnostic decision- making approach applicable to a broad class of membrane separations, quantifies the membrane-area/utility-consumption tradeoff governing total cost, and demonstrates the effectiveness of predictive, estimation-based control for stable and near-optimal operation under uncertainty.