支持向量机在水质评价及水位动态预测中的应用  

Support Vector Machines in Water Quality Assessment and Prediction of Water Level

作　　者：黄鹤[1] 卢文喜[1] 姜雪[1] 

机构地区：[1]吉林大学环境与资源学院,长春130021

出　　处：《中国农村水利水电》2012年第9期30-33,共4页China Rural Water and Hydropower

基　　金：国家自然科学基金资助项目(41072171)

摘　　要：利用支持向量机方法,构建RBF核函数支持向量机水质评价模型和地下水位动态预测模型,对白城地区进行水质评价及地下水位预测。结果表明:白城地区地下水污染严重,以Ⅳ类、Ⅴ类水居多,地下水位动态变化具有周期性。通过实例验证,支持向量机方法评价结果合理,且与综合水质评价法对比分析,更接近实际水质情况;在水位预测中也表现良好,预测准确率达到96.7%。可见该方法在水质评价及水位动态预测中表现出优秀的性能,具有很好的研究价值和推广前景。In this paper, RBF support vector machine model is established for the water quality assessment and prediction of groundwater in Baicheng Area, for the water quality assessment and prediction of water levels. Results indicate that groundwater is seriously polluted in Baieheng, the majority of water quality levels are Ⅳ and Ⅴ, and the groundwater level changes periodically. Through instance validation, the results of support vector machine method for evaluation is reasonable, and are closer to the actual situation of water quality compared with comprehensive water quality assessment method. And it is performed well in the water level forecasl, accuracy rate is 96. 7%. This method in water quality assessment and prediction of water level shows excellent performance, and has a good research value and can be popularized.

关 键 词：支持向量机 水质评价 地下水位动态预测 

分 类 号：X824[环境科学与工程—环境工程]