GONG Qiaoling,GUAN Dongjie.Study on Water Resources Security Evaluation of Chongqing Section of Three Gorges Reservoir Based on BP Neural Network[J].Research of Soil and Water Conservation,2017,24(06):292-299.
基于BP神经网络的三峡库区重庆段水资源安全评价
- Title:
- Study on Water Resources Security Evaluation of Chongqing Section of Three Gorges Reservoir Based on BP Neural Network
- Keywords:
- water resources science; water resources security evaluation; BP neural network; random forests; Three Gorges Reservoir area
- 分类号:
- X52
- 摘要:
- 三峡库区水资源安全关系到库区的生态安全，对库区水资源安全利用进行评价，找出问题较为严重的区域，明确其限制因子，从而为库区水资源可持续发展和水资源安全提供科学的决策依据。结合2000-2014年三峡库区水资源安全数据，选取BP神经网络构建水资源安全利用评价模型，结合ARIMA模型进行指标预测，分析了三峡库区重庆段水资源安全利用的时空差异。研究表明：（1）三峡库区重庆段2014年水资源安全利用总体分布在较不安全到较安全之间，渝中区和大渡口区为较不安全等级，沙坪坝区等7个区县为基本安全，其他区县为较安全等级；在子系统的评价中，主城区的社会安全与供需安全等级均为最低，为极不安全和较不安全等级。（2）2000-2014年水资源安全利用等级主要受生态安全因子制约，供需安全因子次之；2015-2020年水资源安全利用等级主要受供需安全因子限制，生态安全因子次之。2000-2020年水资源安全利用等级总体呈上升趋势。
- Abstract:
- The safety of water resources in the Three Gorges Reservoir area is largely related to the ecological security of the reservoir area. The evaluation on the security utilization of water resources in the Three Gorges Reservoir area facilitates to find out the areas with serious security problems and in turn illustrate the limiting factors that affect the safe use of water resources aiming at providing a scientific basis for the sustainable development and safe utilization of water resources. The water security data of the Three Gorges Reservoir area in 2000-2014 By were chosen, the BP neural network was employed to construct an evaluation mode of safe utilization of water resources. Then, index prediction in combination with an ARIMA model was performed to analyze the temporal and spatial differences of security utilization of water resources in Chongqing section of the Three Gorges Reservoir area. (1) In general, the security utilization of water resources in 2014 was situated in a range from less secure to safer. Specially, Yuzhong and Dadukou Districts were the less safe areas; Shapingba and Jiangbei Districts were the basically safe areas; other districts were the safe areas. With the respect to the evaluated subsystem, the security for both the society and the supply-demand loop in the main urban area were the lowest level. Their security levels were extremely unsafe and unsafe, respectively. (2) During the period from 2000 to 2014, the security level of water resources utilization was mainly constraint from the ecological security factors and successively from the safety factors of supply-demand loop; during the period from 2015 to 2020, the security level of water resources utilization was mainly restricted by the safety factors of supply-demand loop, and followed by the ecological security factors. During the period from 2000 to 2020, the utilization level of water resources security presented the general uptrend.
参考文献/References:
[1] 张凤太,苏维词.贵州省岩溶地区水土资源生态安全研究[J].水电能源科学,2015,33(9):152-154.
[2] Hurk M, Mastenbroek E, Meijerink S. Water safety and spatial development:an institutional comparison between the United Kingdom and the Netherlands[J]. Land Use Policy,2014,36(1):416-426.
[3] Zhou Z Y, Zhang X J, Dong W Y. Fuzzy Comprehensive Evaluation for Safety Guarantee System of Reclaimed[J]. Procedia Environmental Sciences,2013,18:227-235.
[4] 许波,谢谟文,何兴东,等.基于边坡单元的水库库岸稳定性层次分析与评价[J].水土保持通报,2016,36(3):290-296.
[5] 刘洁,谢丽芳,杨国英,等.丰水区城镇化进程与水资源利用的关系:以江苏省为例[J].水土保持通报,2016,36(3):193-199.
[6] 陈亚婷,张阳生,侯红蕊.青海省湟源县滨水环境模糊综合评判及其建设[J].水土保持通报,2013,33(6):187-191.
[7] 明均仁,肖凯.基于R语言的面向需水预测的随机森林方法[J].统计与决策,2012(9):81-83.
[8] 张颖,高倩倩.基于灰色模型和模糊神经网络的综合水质预测模型研究[J].环境工程学报,2015,(2):537-545.
[9] 叶超,郝付军.基于支持向量机和BP神经网络的滑坡变形复合式预测[J].水土保持通报,2016,36(3):332-337.
[10] 崔东文.随机森林回归模型及其在污水排放量预测中的应用[J].供水技术,2014,8(1):31-36.
[11] Xuan W, Quan C, Shuyi L. An optimal water allocation model based on water resources security assessment and its application in Zhangjiakou Region, northern China[J]. Resources, Conservation and Recycling, 2012,69(12):57-65.
[12] 杨家伟.重庆市城镇水安全评价及保障体系研究[D].重庆:重庆大学,2015.
[13] 赖成光,陈晓宏,赵仕威,等.基于随机森林的洪灾风险评价模型及其应用[J].水利学报,2015,46(1):58-66.
[14] 张颖,高倩倩.基于随机森林分类算法的巢湖水质评价[J].环境工程学报,2016,10(2):992-998.
[15] 赵超,申烨红,赵先贵.基于水足迹理论的重庆市2000-2012年水资源评价[J].四川环境,2015,34(3):62-68.
[16] 马骏,李昌晓,魏虹,等.三峡库区生态脆弱性评价[J].生态学报,2015,35(21):7117-7129.
[17] 胡昌军.双隐层BP神经网络模型在区域水安全评价中的应用[J].水资源与水工程学报,2013,24(3):196-200.
[18] 程福亨. BP神经网络与模糊神经网络在空气质量评价中的研究[D].广州:广东工业大学,2012.
[19] 施开放,刁承泰.重庆市三峡库区水资源承载力评价[J].灌溉排水学报,2012,31(5):108-112.
[20] 肖新成,何丙辉,倪九派,等.农业面源污染视角下的三峡库区重庆段水资源的安全性评价:基于DPSIR框架的分析[J].环境科学学报,2013,33(8):2324-2331.
备注/Memo
收稿日期:2016-12-13;改回日期:2017-01-19。
基金项目:国家自然科学基金（41201546）；国家十三五重点研发计划（2016YFC0400700）；重庆市自然科学基金（cstc2012jjA20010）；国家科技支撑计划项目（2014BAB03B01）；重庆交通大学研究科研创新（创新基金）项目（CYS16187）
作者简介:龚巧灵(1991-),女,新疆五家渠人,在读硕士,研究方向:国土资源与3S技术。E-mail:gongqiaoling1437@163.com
通讯作者:官冬杰(1980-),女,黑龙江省富锦人,博士,教授,博士生导师,研究方向:生态环境与可持续发展、水资源安全利用评价、3S应用等。E-mail:guandongjie_2000@163.com