This study investigates the influence of urban heat islands (UHI) on buildings and explores passive design strategies to improve thermo-energy performance while mitigating heat emissions into the urban environment. Concentrating on Latin American cities, the research conducts an intra-urban and climate-related analysis in two scenarios: naturally ventilated and equipped with an HVAC system. The intra-urban analysis considers socioeconomic disparities and diverse urban zones, while the climate-related analysis covers five different cities. This analysis utilizes machine learning models combined with the Non-dominated Sorting Genetic Algorithm II (NSGA-II) for optimization, along with the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) for suitable solution selection. The results reveal potential reductions of energy loads, discomfort hours, and heat emissions, ranging from 2 % to nearly 120 %, depending on the case or scenario evaluated. The Pareto front varieties are discussed, offering design recommendations for addressing housing disparities and optimizing urban zones or cities. The findings suggest that newer building energy standards might underestimate urban warming in historically cold climates. Additionally, shifting toward HVAC use in residential areas could lead to new socioeconomic and environmental issues. This underscores the need for holistic building design that integrates balanced urban planning considerations to mitigate urban overheating.

中文翻译：

---

优化城市住房设计：提高热能性能并减少建筑物的热排放——拉丁美洲案例研究


本研究调查了城市热岛 (UHI) 对建筑物的影响，并探索了被动式设计策略，以提高热能性能，同时减少向城市环境的热量排放。该研究以拉丁美洲城市为重点，在自然通风和配备暖通空调系统两种情况下进行了城市内和气候相关的分析。城市内分析考虑了社会经济差异和不同的城市区域，而气候相关分析则涵盖了五个不同的城市。该分析利用机器学习模型与非支配排序遗传算法 II (NSGA-II) 相结合进行优化，并利用与理想解决方案相似的顺序偏好技术 (TOPSIS) 来选择合适的解决方案。结果显示，能源负荷、不适时间和热量排放可能会减少 2% 到近 120%，具体取决于评估的案例或场景。讨论了帕累托前沿变量，为解决住房差距和优化城市地区或城市提供了设计建议。研究结果表明，新的建筑能源标准可能低估了历史上寒冷气候下的城市变暖。此外，住宅区转向使用暖通空调可能会导致新的社会经济和环境问题。这强调了整体建筑设计的必要性，整合平衡的城市规划考虑因素，以缓解城市过热。