Multi-Objective Optimization of the Energy Efficiency of Rural Dwellings in the Qianbei Region Based on Global Sensitivity Analysis

Based on on-site surveys of rural dwellings in the Qianbei Region, this study addresses the persistent challenge of high energy consumption, poor thermal comfort, and economic constraints. A regionally adaptive design methodology integrating parametric modeling, global sensitivity analysis (GSA), and multi-objective optimization was developed. Using Rhino–Grasshopper, a parametric model with 24 variables was constructed. The Sobol GSA method identified the most influential variables by quantifying both individual and interaction effects. Optimization objectives were annual energy consumption (E), thermal discomfort time (TDT), and life-cycle cost per unit area (LCCunit). Multi-objective optimization using the NSGA-II algorithm shows that the high-performance scenario—150 mm XPS insulation combined with Low-E windows—reduces energy consumption by 52.4% and thermal discomfort hours by 55.4%. The cost-effective scenario, based on an 80 m2 miniaturized design, lowers LCCunit to ¥248,000. The optimal synergistic paradigm integrates a southwest orientation (210°) with XPS insulation and Low-E windows, providing a standardized and replicable approach for rural dwelling construction and retrofit in similar climatic regions.