地理空间分析
计算机科学
聚类分析
节点(物理)
冗余(工程)
分布式计算
数据挖掘
水准点(测量)
空间生态学
蚁群优化算法
理论(学习稳定性)
数据集成
网络分析
比例(比率)
生态学
空间分析
生态网络
大数据
地理
传感器融合
稳健性(进化)
网络拓扑
复杂网络
城市群
城市规划
人工智能
出处
期刊:Land
[Multidisciplinary Digital Publishing Institute]
日期:2026-04-07
卷期号:15 (4): 602-602
被引量:1
摘要
Constructing resilient Ecological Security Patterns (ESPs) in polycentric urban agglomerations is computationally challenging due to persistent scale mismatches between local planning and regional strategies. To address this, we developed a novel Proactive Integration Mechanism (PIM), a computational framework that dynamically optimizes ESPs by algorithmically fusing multi-source geospatial data. The PIM integrates three innovative components: (1) a Function–Structure–Policy data fusion approach that couples Self-Organizing Map clustering of ecosystem services with Morphological Spatial Pattern Analysis and policy data to identify ecological sources; (2) a Dual-Feedback Mechanism that hybridizes circuit theory with an Improved Ant Colony Optimization algorithm for dynamic corridor delineation; and (3) complex network analysis to derive targeted interventions from topological properties. Applied to a node city of the Chengdu-Chongqing Economic Circle, the PIM identified 22 integrated ecological sources and 37 corridors. The optimized network showed enhanced resilience: a deterministic 20.5% increase in circuit redundancy (α-index) and an 8.6% improvement in overall connectivity (γ-index), achieved through minimal topological modifications. Temporal validation (2000–2020) confirmed the high stability of the identified patterns. This study provides a potentially replicable and computationally robust framework that bridges spatial ecology with optimization algorithms, offering a promising paradigm for constructing ESPs in node cities within subtropical urban agglomerations.
科研通智能强力驱动
Strongly Powered by AbleSci AI