Cross-Scale S4ND-UNet for Energy Consumption Optimization in Solvent Recovery

Source abstract

Dynamic solvent-recovery columns are commonly operated with conservative reflux and steam margins because feedcomposition, tray hydraulics, heat inventory, and product-purity requirements vary across time and column position. Thisstudy developed a cross-scale S4ND-UNet surrogate to predict energy demand and identify feasible low-energy operatingconditions. Structured state-space blocks retained long operating memory, while the UNet hierarchy preserved tray-levelpatterns and exchanged them with section- and column-level energy states through gated cross-scale bridges. A constraineddecoder jointly predicted specific reboiler duty, condenser duty, purity margin, and hydraulic feasibility, after which abounded search screened candidate set points. Evaluation used 18,720 one-minute operating windows from acetone-water,ethanol-water, isopropanol-water, and dimethylformamide-water systems. The windows combined simulated tray profileswith pilot-calibrated heat-loss, condenser-approach, and pressure-drop behavior, and the data were separated by operatingcampaign to reduce temporal leakage. The proposed model achieved a heat-duty mean absolute percentage error of 2.31%,compared with 6.42% for temporal convolution and 4.18% for a transformer surrogate. Feasible search reduced meanreboiler duty from 4.53 to 4.12 MJ kg⁻¹, corresponding to 9.04% mean energy saving, while maintaining 98.98 wt% meanproduct purity. Removing the cross-scale bridges increased error to 3.47%, and the complete model stabilized therecommended duty within 18 min, compared with 31 min for the temporal-convolution baseline. These results indicatethat long-memory propagation, multiresolution feature exchange, and feasibility-aware decoding can support energyefficientadvisory operation within the calibrated solvent-recovery envelope.