基于KH-SVM的鄱阳湖水质评价及变化趋势分析

田熊长; 蒋水华; 贾卓; 李琴; 方立东; 张贻林; 肖睿

doi:10.13205/j.hjgc.202301009

基于KH-SVM的鄱阳湖水质评价及变化趋势分析

doi: 10.13205/j.hjgc.202301009

1. 南昌大学工程建设学院, 南昌 330031;
2. 江西省科学院能源研究所, 南昌 330095

基金项目:

江西省水利厅重大科技项目（202224ZDKT24）；江西省科学院省级财政科研项目经费包干制试点示范项目（2021YSBG22013）；国家自然科学基金项目（42202278）

详细信息

作者简介:
田熊长(1996-),男,硕士研究生,主要研究方向为水文水资源与生态水利。1750833725@qq.com

通讯作者:
贾卓(1992-),男,博士,讲师,主要研究方向为水文与水生态环境效应。jiazhuo@ncu.edu.cn

计量
- 文章访问数: 270
- HTML全文浏览量: 46
- PDF下载量: 10
- 被引次数: 0
出版历程
- 收稿日期: 2022-03-10
- 网络出版日期: 2023-03-23

WATER QUALITY EVALUATION AND CHANGE TREND ANALYSIS OF THE POYANG LAKE BASED ON KH-SVM

1. School of Infrastructure Engineering, Nanchang University, Nanchang 330031, China;
2. Institute of Energy Research, Jiangxi Academy of Sciences, Nanchang 330095, China

摘要

摘要: 为了提高水质评价结果的精度，更客观地反映水体环境的质量状况，针对支持向量机（support vector machine，SVM）性能易受模型参数的影响，提出了一种基于磷虾群算法（krill herd，KH）优化SVM的模型，将其与地表水环境质量标准相结合，选取5种代表性水质指标作为模型的输入，5类水质标准作为输出，建立了基于KH-SVM的水质评价模型，并利用该模型对鄱阳湖2013—2018年丰水期、平水期和枯水期的水质状况及变化趋势进行研究。结果表明：鄱阳湖北湖区水质基本维持在GB 3838—2002《地表水环境质量标准》Ⅲ—Ⅳ类，而南湖区水质各时期波动较大，其中丰水期多为Ⅱ类水，平、枯水期以Ⅴ类水为主，整体上2013—2018年鄱阳湖水质恶化趋势没有明显改善，水质级别仍处于非优类水平；构建的KH-SVM模型性能良好，与传统评价方法相比可以较好地实现水质综合评价，能够从整体上更准确、客观地反映水体质量状况。该研究成果可为区域地表水资源科学管理及水环境保护提供科学依据。
- 磷虾群算法 /
- 支持向量机 /
- 水质评价 /
- 鄱阳湖 /
- 参数优化
Abstract: To improve the accuracy of the water quality evaluation results, and reflect the quality status of the water environment more objectively, for the performance of support vector machines (SVM) is susceptible to the model parameters, this study presented an algorithm based on optimized SVM by krill clusters (krill herd, KH), and combined it with the surface water environmental quality standards. Five representative water quality indicators were selected as the input of the model, and five water quality standards were selected as the output. A water quality evaluation model based on KH-SVM as established, and the model was used to study the water quality status and change trend of the Poyang Lake in the high flow period, normal flow period and low flow period from 2013 to 2018. The results showed that the water quality of the Poyang Lake was basically maintained between category Ⅲ and Ⅳ Environment Quality Standard for Surface Water (GB 3838-2002), while it fluctuated greatly in each period, including mostly Ⅱ types and Ⅴ types in the wet season from 2013 to 2018, at the non-optimal level. The KH-SVM model performs well compared with the traditional evaluation method, and can reflect the water quality more accurately and objectively. The research results can provide a scientific basis for the scientific management of regional surface water resources and water environment protection. The results showed that the water quality in the North Lake area of the Poyang Lake was basically maintained between Class Ⅲ and Class Ⅳ, while the water quality in the South Lake area fluctuated greatly in different periods, with Class Ⅱ water in the high flow period and Class Ⅴ water in the normal and low flow periods. On the whole, the deterioration trend of the water quality in the Poyang Lake from 2013 to 2018 had not improved significantly, and the water quality level was still at the non-optimal level. The constructed KH-SVM model had a good performance. Compared with the traditional assessment methods, it could better realize the comprehensive assessment of water quality, and reflect the water quality status as a whole. The research can provide a scientific basis for the scientific management of regional surface water resources and water environment protection.
- krill group algorithm /
- support vector machine /
- water quality evaluation /
- the Poyang Lake /
- parameter optimization

HTML全文

参考文献(26)

[1]	鲁照, 黄河清, 徐力刚, 等. 基于熵权的WQI法在鄱阳湖水质评价中的应用[J]. 环境监测管理与技术, 2021, 33(4):30-34.
[2]	ZHANG Q H, DONG X H, CHEN Y W, et al. Hydrological alterations as the major driver on environmental change in a floodplain Lake Poyang (China):evidence from monitoring and sediment records[J]. Journal of Great Lakes Research, 2018, 44(3):377-387.
[3]	孙文浩, 卞建民, 李一涵, 等. 长白山北坡流域水资源涵养功能与矿泉水开发阈值研究[J]. 水资源与水工程学报, 2022, 33(3):89-97.
[4]	贾卓, 杨国华, 张赫轩, 等. 挠力河流域地下水氮污染特征分析[J].环境污染与防治, 2018, 40(4):418-422 , 434.
[5]	JIA Z, BIAN J M, WANG Y. Impacts of urban land use on the spatial distribution of groundwater pollution, Harbin City, Northeast China[J]. Journal of Contaminant Hydrology, 2018, 215(aug.):29-38.
[6]	丛铭, 阳辉, 张晓静, 等.单因子法与可变模糊法在水质评价中的应用[J].南水北调与水利科技(中英文), 2021, 19(4):720-728.
[7]	李祚泳, 张正健.基于回归支持向量机的指标规范值的水质评价模型[J].中国环境科学, 2013, 33(8):1502-1508.
[8]	李茜, 张建辉, 林兰钰, 等.水环境质量评价方法综述[J].现代农业科技, 2011(19):285-287.
[9]	陈东清, 黄章树, 叶翀. 基于PSO-SVM模型的物流业景气指数组合预测研究[J]. 北京邮电大学学报(社会科学版), 2020, 22(6):52-62.
[10]	尹志勇, 陈永光, 桑博. 基于GA-SVM的电能质量扰动分类方法研究[J]. 河北工业大学学报, 2017, 46(4):75-79.
[11]	李力, 王鑫. 基于KH-SVM的大学生心理健康评价[J]. 微型电脑应用, 2021, 37(5):45-47 , 51.
[12]	毛玉婷, 周晓宇, 王毛兰.枯水期鄱阳湖水体富营养化状态评价[J].南昌大学学报(理科版), 2014, 38(6):596-599.
[13]	江西省水利厅. 江西河湖大典[M]. 武汉:长江出版社, 2010.
[14]	赵志刚, 余德, 韩成云, 等. 2008-2016年鄱阳湖生态经济区生态系统服务价值的时空变化研究[J]. 长江流域资源与环境, 2017, 26(2):198-208.
[15]	刘贺, 张奇, 牛媛媛, 等. 2013-2018年鄱阳湖水环境监测数据集[J]. 中国科学数据(中英文网络版), 2020, 5(2 ):1-7.
[16]	刘新, 许梦文, 赵珍, 等. 鄱阳湖蝶形湖泊水体氮磷等的变化及污染初步评价[J]. 长江流域资源与环境, 2017, 26(8):1189-1198.
[17]	温春云, 刘聚涛, 胡芳, 等. 鄱阳湖水质变化特征及水体富营养化评价[J]. 中国农村水利水电, 2020(11):83-88.
[18]	沙海飞, 范丽丽.地表水溶解氧与化学需氧量、高锰酸盐指数的相关性研究[J].广州化工, 2021, 49(10):105-106 , 110.
[19]	AMIR H G, AMIR H A. Krill herd:a new bio-inspired optimization algorithm[J]. Communications in Nonlinear Science and Numerical Simulation, 2012, 17(12):4831-4845.
[20]	HOFMANN E E, EDWARD HASKELL A G, KCINCK J M, et al. Lagrangian modelling studies of Antarctic krill (Euphausia superba) swarm formation[J]. ICES Journal of Marine Science/Journal du Conseil, 2004, 61:617-631.
[21]	丁成, 王秋萍, 王晓峰. 基于广义反向学习的磷虾群算法及其在数据聚类中的应用[J]. 计算机应用, 2019, 39(2):336-342.
[22]	OLIVIER C, PATRICK H, VLADIMIR V, et al. Support vector machines for histogram-based image classification[J]. IEEE Transactions on Neural Networks, 1999, 10(5):1055-1055.
[23]	KANESAMOORTHY K, DISSANAYAKE M B. Prediction of treatment failure of tuberculosis using support vector machine with genetic algorithm[J]. International Journal of Mycobacteriology, 2021, 10(3):279-284.
[24]	郑丙辉, 刘琰. 地表水环境质量标准修订的必要性及其框架设想[J]. 环境保护, 2014, 42(20):39-41.
[25]	GARCÍ-NIETO P J, GARCÍA-GONZALO E, SÁNCHEZ-LASHERAS F J, et al. Hybrid PSO-SVM-based method for forecasting of the remaining useful life for aircraft engines and evaluation of its reliability[J]. Reliability Engineering & System Safety, 2015, 138(Jun.):219-231.
[26]	MEHMET A T K, HUSEYIN S. Parameters selection for water quality index in the assessment of the environmental impacts of land-based trout farms[J]. Ecological Indicators:Integrating, Monitoring, Assessment and Management, 2014, 36(Jan.):672-681.