期刊
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
卷 7, 期 2, 页码 -出版社
MDPI
DOI: 10.3390/ijgi7020063
关键词
land-use allocation; multi-objective; spatial optimization; Pareto; artificial bee colony algorithm
资金
- Open Research Fund of Key Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences [2015LDE002]
- Director Foundation of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences [Y6SJ2300CX]
Land-use allocation is of great significance in urban development. This type of allocation is usually considered to be a complex multi-objective spatial optimization problem, whose optimized result is a set of Pareto-optimal solutions (Pareto front) reflecting different tradeoffs in several objectives. However, obtaining a Pareto front is a challenging task, and the Pareto front obtained by state-of-the-art algorithms is still not sufficient. To achieve better Pareto solutions, taking the grid-representative land-use allocation problem with two objectives as an example, an artificial bee colony optimization algorithm for multi-objective land-use allocation (ABC-MOLA) is proposed. In this algorithm, the traditional ABC's search direction guiding scheme and solution maintaining process are modified. In addition, a knowledge-informed neighborhood search strategy, which utilizes the auxiliary knowledge of natural geography and spatial structures to facilitate the neighborhood spatial search around each solution, is developed to further improve the Pareto front's quality. A series of comparison experiments (a simulated experiment with small data volume and a real-world data experiment for a large area) shows that all the Pareto fronts obtained by ABC-MOLA totally dominate the Pareto fronts by other algorithms, which demonstrates ABC-MOLA's effectiveness in achieving Pareto fronts of high quality.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据