【摘要】 城市社区的空间布局对居民生活质量和城市的可持续发展至关重要，尤其在实现空间公平方面具有重要意义。传统的城市规划方法往往过于依赖专家经验，难以全面考虑各类需求，导致空间资源的分配不均和服务获取不公平。为此，文章基于“15分钟生活圈”理念，提出了结合深度强化学习（DRL）算法的城市空间规划方法，通过优化绿色空间的均衡性指标与圈层式的服务配置指标，实现了自动化的空间布局优化。实验结果表明，该方法能够生成较好效率的空间布局，并能够灵活响应不同社区的需求。更多还原

【Abstract】 The spatial layout of urban communities is vital to residents’ quality of life and the sustainable development of cities, especially for promoting spatial equity. Traditional urban planning approaches often rely heavily on expert experience, making it difficult to comprehensively address diverse needs and easily leading to imbalanced allocation of spatial resources and inequitable service access. In response, this study proposes an intelligent urban land use planning approach integrating the concept of the "15-minute urban life circle" with a Deep Reinforcement Learning(DRL)algorithm. By optimizing indicators of green space equity and hierarchical service provision, the approach enables automated optimization of spatial layouts. Experimental results show that this method can efficiently generate optimized spatial configurations and flexibly adapt to the needs of different communities.更多还原