超球支持向量机的地下水水质参数自动监测技术研究  

Research on Automatic Monitoring Technology of Groundwater Quality Parameters based on Hypersphere Support Vector Machine

作　　者：孙英军 SUN Yingjun(Zhejiang Hydrologic Management Center, Hangzhou 310000, China)

机构地区：[1]浙江省水文管理中心,浙江杭州310000

出　　处：《中国水能及电气化》2021年第10期45-49,共5页China Water Power & Electrification

摘　　要：水质参数指标项相对较多,存在信息重叠等问题,导致水质参数监测精度降低,为此,提出基于超球支持向量机的地下水水质参数自动监测技术。通过主成分分析法消除信息重叠的影响,引入改进粒子群法,通过自适应调节惯性权重的方式,对动态变化学习因子进行融合,优化超球支持向量机模型运行参数,利用优化后的支持向量机对水质等级进行评价,构建评价指标体系,并将该体系作为建立模型训练样本集合的基础,可降低人为因素与监测数据对模型运行的干扰,完成自动监测与报警。依托地下水段,将其作为测试目标进行水质参数自动监测。监测结果表明,不同测试点测量结果与实际结果相同,表明基于超球支持向量机的地下水水质参数自动监测技术监测性能较高。There are many water quality parameter indexes.Overlapping information and other problems are available,thereby resulting in the decrease of the monitoring accuracy of water quality parameters.Therefore,an automatic monitoring technology of groundwater water quality parameters based on hypersphere support vector machine is proposed.The principal component analysis(PCA)is adopted to eliminate the influence of overlapping information.The improved particle swarm method is introduced.The dynamic change learning factors are integrated to optimize operation parameters of the hypersphere support vector machine(SVM)modelby means of an adaptive adjustment of inertia weight.The optimized support vector machine(SVM)is utilized to evaluate the water quality grade and construct an evaluation index system.The system is regarded as the basis for establishing the model training sample set,thereby reducing the interference of human factors and monitoring data to the model operation,and completing the automatic monitoring and alarm.An underwater section is adopted as the test objective for automatically monitoring water quality parameters.The monitoring result shows that the measurement results of different test points are the same as the actual results.It is obvious that the underground water quality parameter automatic monitoring technology based on the hypersphere support vector machine has better monitoring performance.

关 键 词：地下水 超球支持向量机 水质参数 主成分分析 自动监测 

分 类 号：X832[环境科学与工程—环境工程]