期刊文献+

基于支持向量机理论的地下水水源地动态预测模型 被引量:1

Groundwater Source Dynamic Prediction Model Based on Support Vector Machines Theory
原文传递
导出
摘要 盘锦市盘东地下水水源地是辽河油田工业和生活用水主要水源,长期观测资料分析表明开采量是影响水源地水位动态的主要因素,二者之间存在着显著的相关性。该文在介绍支持向量机基本原理和实现算法的基础上,探讨了如何合理选择核函数,建立支持向量机回归方法水源地水位动态预测模型。通过对比多项式核函数和径向基函数(RBF)核函数的训练和模拟结果,发现后者预测精度高。研究成果表明,相对于传统的回归分析方法,基于支持向量机回归方法的动态模型的模拟预测能力明显提高。该文中的动态模型方法在地下水位动态预测中具有广泛的应用前景。 Pandong groundwater source in Panjin city is the main water source of industry and life for Liaohe oil base, through analyzing the long term observation data, the exploitation quantity is the main factors that affecting the groundwater level dynamic regime of water source, There is a significant correlation between the groundwater level and the exploitation quantity. Based on the principles of support vector machines and implementation algorithm, how to select kernel functions reasonably and establish the groundwater level prediction model are discussed by using support vector machines regression method. Through the results comparison between polynomial kernel function and radial basis function, it shows the latter has high accuracy. The analysis results show that the simulation predicted ability of dynamic model method has improved obviously comparing with traditional regression method. The proposed dynamic model method has wide application prospect in dynamic prediction of groundwater regime.
作者 刘长生
出处 《勘察科学技术》 2010年第1期22-25,共4页 Site Investigation Science and Technology
关键词 水源地 动态 支持向量机 预测 water source dynamic regime support vector machines predication
  • 相关文献

参考文献9

二级参考文献48

共引文献187

同被引文献7

引证文献1

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部