Novel integrated meta-sampling approach for smart building design: real-time data analytics and energy performance optimization practice

Atrium design significantly influences the energy performance of public buildings, especially in climate-sensitive regions. This study develops a parametric optimization framework integrating automated modeling, large-scale simulation, and meta-sample analysis to evaluate the impact of eight atrium design parameters on energy use. A benchmark model library of 39,202 atrium configurations was constructed to assess heating, cooling, and lighting energy consumption. Sensitivity analysis revealed that atrium area and internal openness are the most influential factors for cooling and heating demand. South-facing atriums consistently showed lower annual cooling energy demand (CEUI) and annual heating energy demand (HEUI). One actionable strategy is reducing internal openness (Int) and increasing external openness (Out) to significantly lower energy use. Multi-objective optimization demonstrated up to 40.29 % reduction in annual total energy demand (EUI). The meta-sample model, validated against full-sample statistics, achieved 87.6 % computational savings with maximum output error of only 0.35 %. These findings provide quantitative design guidance for early-stage atrium optimization and enable fast integration with parametric BIM workflows.