PSOLA: A Heuristic Land-Use Allocation Model Using Patch-Level Operations and Knowledge-Informed Rules.
PSOLA: A Heuristic Land-Use Allocation Model Using Patch-Level Operations and Knowledge-Informed Rules.
复制标题
PSOLA：使用补丁级操作和知识告知规则的启发式土地利用分配模型
DOI：
10.1371/journal.pone.0157728
复制
发表时间：
2016
期刊：
影响因子：
3.7
通讯作者：
Liu Y
中科院分区：
文献类型：
--
作者：
Liu Y;Peng J;Jiao L;Liu Y
土地利用优化配置促进了公共服务的社会公平性，提高了土地利用活动的经济效益，降低了土地利用规划的生态风险，对区域可持续发展具有重要意义。大多数土地利用优化模型使用单元级操作来分配土地利用，该操作使土地利用斑块碎片化。这些模型不能很好地与土地利用规划知识，导致不合理的土地利用模式。本研究以粒子群最佳化法建构一启发式土地使用分配模式。该模型以斑块级操作分配土地使用，以避免碎片化。斑块层次的操作包括斑块边缘算子、斑块大小算子和斑块紧密度算子，它们约束土地利用斑块的大小和形状。该模型还集成了知识知情的规则，以提供辅助知识的土地利用规划优化过程中。知识型规则包括适宜性、可达性、土地利用政策和利益相关者偏好。为了验证PSOLA模型，在浙江省高桥镇，中国进行了案例研究。结果表明，PSOLA模型的社会效益、经济效益、生态效益和综合效益分别比基本粒子群优化算法提高了3.60%、7.10%、1.53%和4.06%，验证了改进的有效性。此外，该模型有一个开放的架构，使其扩展为一个通用的工具，以支持决策的土地利用规划。
Optimizing land-use allocation is important to regional sustainable development, as it promotes the social equality of public services, increases the economic benefits of land-use activities, and reduces the ecological risk of land-use planning. Most land-use optimization models allocate land-use using cell-level operations that fragment land-use patches. These models do not cooperate well with land-use planning knowledge, leading to irrational land-use patterns. This study focuses on building a heuristic land-use allocation model (PSOLA) using particle swarm optimization. The model allocates land-use with patch-level operations to avoid fragmentation. The patch-level operations include a patch-edge operator, a patch-size operator, and a patch-compactness operator that constrain the size and shape of land-use patches. The model is also integrated with knowledge-informed rules to provide auxiliary knowledge of land-use planning during optimization. The knowledge-informed rules consist of suitability, accessibility, land use policy, and stakeholders’ preference. To validate the PSOLA model, a case study was performed in Gaoqiao Town in Zhejiang Province, China. The results demonstrate that the PSOLA model outperforms a basic PSO (Particle Swarm Optimization) in the terms of the social, economic, ecological, and overall benefits by 3.60%, 7.10%, 1.53% and 4.06%, respectively, which confirms the effectiveness of our improvements. Furthermore, the model has an open architecture, enabling its extension as a generic tool to support decision making in land-use planning.