【摘要】 针对传统优化模型难以将土地利用数量结构与空间布局优化有效统一的问题,本文提出基于混合蛙跳算法的土地利用优化模型,以地理栅格为基本操作单元,引入首尾排序分组、智能学习算子与变异算子改进算法,实现土地利用空间格局优化。以重庆市渝北区2016年的土地利用数据对2030年土地利用空间格局进行优化,优化前生态系统服务价值为2.012×10~9元,优化后为2.099×10~9元,区域经济总产出优化前为1.263×1011元,优化后为2.148×1011元,土地利用集约度优化前为0.654,优化后为0.812,增长幅度分别为4.3%、70.1%、26.3%。研究表明利用混合蛙跳算法建立土地利用优化模型,能够在多个优化目标与限制条件下,同时进行土地利用数量与空间格局优化,具有较强的全局寻优能力与较快的收敛速度。更多还原

【Abstract】 In traditional models,there has been difficulty in effectively optimization of both quantity structure and spatial pattern for land use. This paper put forward a land use optimization model by Shuffled Frog Leaping Algorithm( SFLA). This model was constructed based on geographic grid cells,with a head-to-tail sorting group method,intelligent learning and mutation operators were introduced to improve the algorithm,and the Yubei district of Chongqing city was taken as a case study. The land use spatial pattern of study area in 2030 was optimized by the model,with its land use data in 2016. The value of ecosystem services in this region was 2. 012 × 10~9 Yuan before optimization,whereas the number increased to 2. 099× 10~9 Yuan after the optimization. The gross economic output before and after optimization was 1. 263 × 1011 Yuan and 2. 148 × 1011 Yuan respectively. The land use intensity before and after optimization was 0. 654 and 0. 812 respectively. The growth rates of ecosystem services value,gross economic output and land use intensity were 4. 3%,70. 1% and 26. 3% respectively. The result showed that land use optimization model based on the SFLA could effectively optimize the quantity structure and spatial pattern of land use at the same time under several optimization objectives and restrictive factors. This method was proved to possess strong global optimization ability and fast convergence.更多还原