Answers are grounded in this paper and cite their sources.
New Mongolian yurt prototypes require coordinated energy, daylight, and thermal-comfort decisions. This study evaluated height, polygon edge number, opening configuration, and envelope system using 336 parametric designs. Held-out R2 values were 0.9549, 0.9947, and 0.9705 for annual energy, mean illuminance (Mean Lux), and predicted percentage dissatisfied (PPD), respectively. SHAP interpretation was examined alongside construction properties and matched simulations. Across 84 matched S3-to-S2 substitutions, mean casewise energy and PPD decreased by 22.96% and 15.79%. Screening 9,180 archived candidates identified 603 non-dominated inputs representing 24 distinct objective vectors. Against a separately simulated conventionalized-envelope baseline, the selected compromise predicted 32.89% lower energy and 38.79% higher Mean Lux; this comparison combines geometry, opening, and envelope changes. A supplementary calculation implies 22.58% lower energy, with a 13.31% discrepancy under incompletely matched settings. PPD increased by 0.81 percentage points, approximately 3.24 times the reported held-out MAE, indicating a directional comfort cost rather than a resolved difference. Controlled E06-to-E08 predictions quantified design-translation consequences. The workflow supports early-stage comparison; matched simulation, spatial daylight assessment, and evaluation of the final geometry remain necessary.
No takes yet. Share an insight, caveat, or question.
Yu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: