水资源空间优化配置的群智能算法改进与仿真
作者简介:侯景伟(1973-),男,博士,研究方向为水资源时空优化配置、算法设计、GIS开发和应用等。E-mail:hjwei2005@163.com
收稿日期: 2013-10-08
要求修回日期: 2013-11-09
网络出版日期: 2015-04-10
基金资助
宁夏自然科学基金重点项目(NZ14002)
Improvement and Simulation of Swarm Intelligence Algorithm for Spatial Optimal Allocation for Water Resources
Received date: 2013-10-08
Request revised date: 2013-11-09
Online published: 2015-04-10
Copyright
本文尝试用群智能算法中的Pareto蚁群算法(PACA)求解复杂的水资源空间优化配置问题。首先,建立了以社会、经济和生态综合效益最大的目标函数,以水质、需水和供水为约束条件的水资源空间优化配置模型,并采用局部信息素强度限制,全局信息素动态更新等策略,对PACA进行改进,使蚂蚁向信息素浓度大的优化边界移动,以提高PACA的全局搜索能力和收敛速度。本文以河南省镇平县为仿真对象,借助RS和GIS,利用改进的PACA求解水资源空间优化配置模型,得到地表水、地下水、外调水的最优配置方案和最佳经济、社会、生态效益方案。通过对PACA性能指标的分析,以及对PACA改进前后解的寻优对比,表明了PACA经过改进后能有效地求解多目标、大规模的水资源空间优化配置模型,提高了寻优性能、收敛速度和全局搜索能力。
关键词: 优化配置; 水资源; 遥感; 地理信息系统; Pareto蚁群算法
侯景伟 , 吴建军 . 水资源空间优化配置的群智能算法改进与仿真[J]. 地球信息科学学报, 2015 , 17(4) : 431 -437 . DOI: 10.3724/SP.J.1047.2015.00431
In order to solve spatial optimal allocation problem of water resource with multi-objective functions and multi-constrained conditions, Pareto ant colony algorithm (PACA) is used in this study. The model for spatial optimal allocation of water resources is established. Its objective function is the largest benefits from economy, society and environment. And its constraints include water supply, water demand and water quality. PACA is improved according to such strategies as limiting local pheromone scope and dynamically updating global pheromone. Then, GIS software is developed with the help of VB. NET 2008, ArcGIS Engine and Access. Zhenping County, Henan Province, China is selected as a study area. Data about water resources in the study area are handled using RS and GIS technology. The model is solved with PACA in the GIS environment. Spatial optimal allocation schemes of water resources, including surface water, groundwater and transfer water, are obtained. And spatial optimal benefit schemes of water resources, including economic, social and ecological benefits are also obtained. The optimal results obtained from PACA are compared with other intelligent optimization algorithms. Robustness performance, optimal performance and time performance of the improved PACA are 5.38, 0.398 and 21.6, respectively. The three performances of the ACA, however, are 8.16, 2.108 and 36.8, respectively. The results indicate that the integration of RS, GIS and PACA can effectively improve the performance of large-scale, multi-objective optimization model of water resources. This method can enhance the global search capability, the convergence speed and the result’s precision.
Key words: optimal allocation; water resources; RS; GIS; Pareto Ant Colony Algorithm (PACA)
Fig.1 Natural conditions in Zhenping County图1 镇平县自然概况 |
Fig.2 Spatial distribution of water demand in Zhenping County图2 镇平县需水量空间分布(m3/ pixel.a) |
Fig.3 Interface of optimal allocation for water resources图3 水资源优化配置的界面 |
Fig.4 Optimal spatial allocation schemes图4 最佳空间优化配置方案(m3/ pixel.a) |
Fig.5 Optimal benefit schemes图5 最佳效益方案 |
Tab.1 Performance validation of the improved PACA表1 改进PACA的性能验证 |
算法类型 | 鲁棒性能指标 | 最佳性能指标 | 时间性能指标 |
---|---|---|---|
改进PACA | 5.38 | 0.397 | 21.6 |
基本蚁群算法 | 8.16 | 2.108 | 36.8 |
Fig.6 Optimal contrast of the solutions before and after improvement of PACA图6 PACA改进前、后解的寻优对比 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
|
[31] |
|
/
〈 | 〉 |